Fix agent session tests to mock Memory component

CITATION.cff

@@ -1,55 +0,0 @@
-cff-version: 1.2.0
-message: "If you use this software, please cite it using the following metadata."
-title: "OpenHands: An Open Platform for AI Software Developers as Generalist Agents"
-authors:
-  - family-names: Wang
-    given-names: Xingyao
-  - family-names: Li
-    given-names: Boxuan
-  - family-names: Song
-    given-names: Yufan
-  - family-names: Xu
-    given-names: Frank F.
-  - family-names: Tang
-    given-names: Xiangru
-  - family-names: Zhuge
-    given-names: Mingchen
-  - family-names: Pan
-    given-names: Jiayi
-  - family-names: Song
-    given-names: Yueqi
-  - family-names: Li
-    given-names: Bowen
-  - family-names: Singh
-    given-names: Jaskirat
-  - family-names: Tran
-    given-names: Hoang H.
-  - family-names: Li
-    given-names: Fuqiang
-  - family-names: Ma
-    given-names: Ren
-  - family-names: Zheng
-    given-names: Mingzhang
-  - family-names: Qian
-    given-names: Bill
-  - family-names: Shao
-    given-names: Yanjun
-  - family-names: Muennighoff
-    given-names: Niklas
-  - family-names: Zhang
-    given-names: Yizhe
-  - family-names: Hui
-    given-names: Binyuan
-  - family-names: Lin
-    given-names: Junyang
-  - family-names: Brennan
-    given-names: Robert
-  - family-names: Peng
-    given-names: Hao
-  - family-names: Ji
-    given-names: Heng
-  - family-names: Neubig
-    given-names: Graham
-year: 2024
-doi: "10.48550/arXiv.2407.16741"
-url: "https://arxiv.org/abs/2407.16741"

ISSUE_TRIAGE.md

@@ -2,13 +2,12 @@
 These are the procedures and guidelines on how issues are triaged in this repo by the maintainers.
 
 ## General
-* All issues must be tagged with **enhancement**, **bug** or **troubleshooting/help**.
-* Issues may be tagged with what it relates to (**agent quality**, **frontend**, **resolver**, etc.).
+* Most issues must be tagged with **enhancement** or **bug**.
+* Issues may be tagged with what it relates to (**backend**, **frontend**, **agent quality**, etc.).
 
 ## Severity
 * **Low**: Minor issues or affecting single user.
 * **Medium**: Affecting multiple users.
-* **High**: High visibility issues or affecting many users.
 * **Critical**: Affecting all users or potential security issues.
 
 ## Effort
@@ -19,14 +18,8 @@ These are the procedures and guidelines on how issues are triaged in this repo b
 
 ## Not Enough Information
 * User is asked to provide more information (logs, how to reproduce, etc.) when the issue is not clear.
-* If an issue is unclear and the author does not provide more information or respond to a request,
-the issue may be closed as **not planned** (Usually after a week).
+* If an issue is unclear and the author does not provide more information or respond to a request, the issue may be closed as **not planned** (Usually after a week).
 
 ## Multiple Requests/Fixes in One Issue
 * These issues will be narrowed down to one request/fix so the issue is more easily tracked and fixed.
 * Issues may be broken down into multiple issues if required.
-
-## Stale and Auto Closures
-* In order to keep a maintainable backlog, issues that have no activity within 30 days are automatically marked as **Stale**.
-* If issues marked as **Stale** continue to have no activity for 7 more days, they will automatically be closed as not planned.
-* Issues may be reopened by maintainers if deemed important.

config.template.toml

@@ -95,11 +95,6 @@ workspace_base = "./workspace"
 # List of allowed file extensions for uploads
 #file_uploads_allowed_extensions = [".*"]
 
-# Whether to enable the default LLM summarizing condenser when no condenser is specified in config
-# When true, a LLMSummarizingCondenserConfig will be used as the default condenser
-# When false, a NoOpCondenserConfig (no summarization) will be used
-#enable_default_condenser = true
-
 #################################### LLM #####################################
 # Configuration for LLM models (group name starts with 'llm')
 # use 'llm' for the default LLM config
@@ -299,69 +294,6 @@ llm_config = 'gpt3'
 # The security analyzer to use (For Headless / CLI only -  In Web this is overridden by Session Init)
 #security_analyzer = ""
 
-#################################### Condenser #################################
-# Condensers control how conversation history is managed and compressed when
-# the context grows too large. Each agent uses one condenser configuration.
-##############################################################################
-[condenser]
-# The type of condenser to use. Available options:
-# - "noop": No condensing, keeps full history (default)
-# - "observation_masking": Keeps full event structure but masks older observations
-# - "recent": Keeps only recent events and discards older ones
-# - "llm": Uses an LLM to summarize conversation history
-# - "amortized": Intelligently forgets older events while preserving important context
-# - "llm_attention": Uses an LLM to prioritize most relevant context
-type = "noop"
-
-# Examples for each condenser type (uncomment and modify as needed):
-
-# 1. NoOp Condenser - No additional settings needed
-#type = "noop"
-
-# 2. Observation Masking Condenser
-#type = "observation_masking"
-# Number of most-recent events where observations will not be masked
-#attention_window = 100
-
-# 3. Recent Events Condenser
-#type = "recent"
-# Number of initial events to always keep (typically includes task description)
-#keep_first = 1
-# Maximum number of events to keep in history
-#max_events = 100
-
-# 4. LLM Summarizing Condenser
-#type = "llm"
-# Reference to an LLM config to use for summarization
-#llm_config = "condenser"
-# Number of initial events to always keep (typically includes task description)
-#keep_first = 1
-# Maximum size of history before triggering summarization
-#max_size = 100
-
-# 5. Amortized Forgetting Condenser
-#type = "amortized"
-# Number of initial events to always keep (typically includes task description)
-#keep_first = 1
-# Maximum size of history before triggering forgetting
-#max_size = 100
-
-# 6. LLM Attention Condenser
-#type = "llm_attention"
-# Reference to an LLM config to use for attention scoring
-#llm_config = "condenser"
-# Number of initial events to always keep (typically includes task description)
-#keep_first = 1
-# Maximum size of history before triggering attention mechanism
-#max_size = 100
-
-# Example of a custom LLM configuration for condensers that require an LLM
-# If not provided, it falls back to the default LLM
-#[llm.condenser]
-#model = "gpt-4o"
-#temperature = 0.1
-#max_tokens = 1024
-
 #################################### Eval ####################################
 # Configuration for the evaluation, please refer to the specific evaluation
 # plugin for the available options

docs/modules/usage/configuration-options.md

@@ -385,11 +385,6 @@ To use these with the docker command, pass in `-e SANDBOX_<option>`. Example: `-
   - Default: `false`
   - Description: Use host network
 
-- `runtime_binding_address`
-  - Type: `str`
-  - Default: `127.0.0.1`
-  - Description: The binding address for the runtime ports.  It specifies which network interface on the host machine Docker should bind the runtime ports to.
-
 ### Linting and Plugins
 - `enable_auto_lint`
   - Type: `bool`

docs/modules/usage/runtimes.md

@@ -84,36 +84,3 @@ docker run # ...
     -e MODAL_API_TOKEN_ID="your-id" \
     -e MODAL_API_TOKEN_SECRET="your-secret" \
 ```
-
-## Daytona Runtime
-
-Another option is using [Daytona](https://www.daytona.io/) as a runtime provider:
-
-### Step 1: Retrieve Your Daytona API Key
-1. Visit the [Daytona Dashboard](https://app.daytona.io/dashboard/keys).
-2. Click **"Create Key"**.
-3. Enter a name for your key and confirm the creation.
-4. Once the key is generated, copy it.
-
-### Step 2: Set Your API Key as an Environment Variable
-Run the following command in your terminal, replacing `<your-api-key>` with the actual key you copied:
-```bash
-export DAYTONA_API_KEY="<your-api-key>"
-```
-
-This step ensures that OpenHands can authenticate with the Daytona platform when it runs.
-
-### Step 3: Run OpenHands Locally Using Docker
-To start the latest version of OpenHands on your machine, execute the following command in your terminal:
-```bash
-bash -i <(curl -sL https://get.daytona.io/openhands)
-```
-
-#### What This Command Does:
-- Downloads the latest OpenHands release script.
-- Runs the script in an interactive Bash session.
-- Automatically pulls and runs the OpenHands container using Docker.
-
-Once executed, OpenHands should be running locally and ready for use.
-
-For more details and manual initialization, view the entire [README.md](https://github.com/All-Hands-AI/OpenHands/blob/main/openhands/runtime/impl/daytona/README.md)

docs/src/components/Demo/Demo.tsx

@@ -5,6 +5,9 @@ export function Demo() {
   const videoRef = React.useRef<HTMLVideoElement>(null);
 
   return (
+    <div
+      style={{ paddingBottom: "10px", paddingTop: "10px", textAlign: "center" }}
+    >
       <video
         playsInline
         autoPlay={true}
@@ -17,5 +20,6 @@ export function Demo() {
       >
         <source src="img/teaser.mp4" type="video/mp4"></source>
       </video>
+    </div>
   );
 }

docs/src/components/Demo/index.module.css

@@ -1,5 +1,6 @@
 .demo {
   width: 100%;
+  padding: 30px;
   max-width: 800px;
   text-align: center;
   border-radius: 40px;

docs/src/components/HomepageHeader/HomepageHeader.tsx

@@ -17,29 +17,6 @@ export function HomepageHeader() {
 
         <p className="header-subtitle">{siteConfig.tagline}</p>
 
-        <div style={{
-          textAlign: 'center',
-          fontSize: '1.2rem',
-          maxWidth: '800px',
-          margin: '0 auto',
-          padding: '0rem 0rem 1rem'
-        }}>
-        <p style={{ margin: '0' }}>
-          Use AI to tackle the toil in your backlog. Our agents have all the same tools as a human developer: they can modify code, run commands, browse the web,
-          call APIs, and yes-even copy code snippets from StackOverflow.
-          <br/>
-          <Link to="https://docs.all-hands.dev/modules/usage/installation"
-            style={{
-              textDecoration: 'underline',
-              display: 'inline-block',
-              marginTop: '0.5rem'
-            }}
-          >
-            Get started with OpenHands.
-          </Link>
-        </p>
-      </div>
-
         <div align="center" className="header-links">
           <a href="https://github.com/All-Hands-AI/OpenHands/graphs/contributors"><img src="https://img.shields.io/github/contributors/All-Hands-AI/OpenHands?style=for-the-badge&color=blue" alt="Contributors" /></a>
           <a href="https://github.com/All-Hands-AI/OpenHands/stargazers"><img src="https://img.shields.io/github/stars/All-Hands-AI/OpenHands?style=for-the-badge&color=blue" alt="Stargazers" /></a>
@@ -50,9 +27,12 @@ export function HomepageHeader() {
           <a href="https://discord.gg/ESHStjSjD4"><img src="https://img.shields.io/badge/Discord-Join%20Us-purple?logo=discord&logoColor=white&style=for-the-badge" alt="Join our Discord community" /></a>
           <a href="https://github.com/All-Hands-AI/OpenHands/blob/main/CREDITS.md"><img src="https://img.shields.io/badge/Project-Credits-blue?style=for-the-badge&color=FFE165&logo=github&logoColor=white" alt="Credits" /></a>
           <br/>
+          <a href="https://docs.all-hands.dev/modules/usage/getting-started"><img src="https://img.shields.io/badge/Documentation-000?logo=googledocs&logoColor=FFE165&style=for-the-badge" alt="Check out the documentation" /></a>
           <a href="https://arxiv.org/abs/2407.16741"><img src="https://img.shields.io/badge/Paper%20on%20Arxiv-000?logoColor=FFE165&logo=arxiv&style=for-the-badge" alt="Paper on Arxiv" /></a>
           <a href="https://huggingface.co/spaces/OpenHands/evaluation"><img src="https://img.shields.io/badge/Benchmark%20score-000?logoColor=FFE165&logo=huggingface&style=for-the-badge" alt="Evaluation Benchmark Score" /></a>
         </div>
+
+        <Demo />
       </div>
     </div>
   );

docs/src/css/homepageHeader.css

@@ -1,14 +1,14 @@
 /* homepageHeader.css */
 
 .homepage-header {
-  padding: 1rem 0;
+  height: 800px;
 }
 
 .header-content {
   display: flex;
   flex-direction: column;
   align-items: center;
-  padding: 1rem;
+  padding: 2rem;
   font-weight: 300;
   width: 100%;
 }
@@ -25,7 +25,6 @@
 
 .header-subtitle {
   font-size: 1.5rem;
-  margin: 0.5rem 0;
 }
 
 .header-links {

docs/src/pages/index.tsx

@@ -2,7 +2,15 @@ import useDocusaurusContext from '@docusaurus/useDocusaurusContext';
 import Layout from '@theme/Layout';
 import { HomepageHeader } from '../components/HomepageHeader/HomepageHeader';
 import { translate } from '@docusaurus/Translate';
-import { Demo } from "../components/Demo/Demo";
+
+export function Header({ title, summary }): JSX.Element {
+  return (
+    <div>
+      <h1>{title}</h1>
+      <h2 style={{ fontSize: '3rem' }}>{summary}</h2>
+    </div>
+  );
+}
 
 export default function Home(): JSX.Element {
   const { siteConfig } = useDocusaurusContext();
@@ -15,14 +23,11 @@ export default function Home(): JSX.Element {
       })}
     >
       <HomepageHeader />
-
-      <div style={{ textAlign: 'center', padding: '1rem 0' }}>
-        <Demo />
-      </div>
-
-      <div style={{ textAlign: 'center', padding: '0.5rem 2rem 1.5rem' }}>
+      <div style={{ textAlign: 'center', padding: '2rem' }}>
+        <br />
         <h2>Most Popular Links</h2>
         <ul style={{ listStyleType: 'none'}}>
+          <li><a href="/modules/usage/Installation">How to Run OpenHands</a></li>
           <li><a href="/modules/usage/prompting/microagents-repo">Customizing OpenHands to a repository</a></li>
           <li><a href="/modules/usage/how-to/github-action">Integrating OpenHands with Github</a></li>
           <li><a href="/modules/usage/llms#model-recommendations">Recommended models to use</a></li>

evaluation/benchmarks/EDA/run_infer.py

@@ -24,7 +24,6 @@ from openhands.core.config import (
 from openhands.core.logger import openhands_logger as logger
 from openhands.core.main import create_runtime, run_controller
 from openhands.events.action import MessageAction
-from openhands.utils.async_utils import call_async_from_sync
 
 game = None
 
@@ -122,7 +121,6 @@ def process_instance(
 
     # Here's how you can run the agent (similar to the `main` function) and get the final task state
     runtime = create_runtime(config)
-    call_async_from_sync(runtime.connect)
 
     state: State | None = asyncio.run(
         run_controller(

evaluation/benchmarks/agent_bench/run_infer.py

@@ -34,7 +34,6 @@ from openhands.core.main import create_runtime, run_controller
 from openhands.events.action import AgentFinishAction, CmdRunAction, MessageAction
 from openhands.events.observation import CmdOutputObservation
 from openhands.runtime.base import Runtime
-from openhands.utils.async_utils import call_async_from_sync
 
 
 def get_config(
@@ -211,7 +210,6 @@ def process_instance(
     # =============================================
 
     runtime: Runtime = create_runtime(config)
-    call_async_from_sync(runtime.connect)
 
     initialize_runtime(runtime, instance=instance)
 

evaluation/benchmarks/aider_bench/run_infer.py

@@ -34,7 +34,6 @@ from openhands.core.main import create_runtime, run_controller
 from openhands.events.action import CmdRunAction, MessageAction
 from openhands.events.observation import CmdOutputObservation
 from openhands.runtime.base import Runtime
-from openhands.utils.async_utils import call_async_from_sync
 
 # Configure visibility of unit tests to the Agent.
 USE_UNIT_TESTS = os.environ.get('USE_UNIT_TESTS', 'false').lower() == 'true'
@@ -204,7 +203,7 @@ def process_instance(
     # =============================================
 
     runtime: Runtime = create_runtime(config)
-    call_async_from_sync(runtime.connect)
+
     initialize_runtime(runtime, instance=instance)
 
     # Here's how you can run the agent (similar to the `main` function) and get the final task state

evaluation/benchmarks/biocoder/run_infer.py

@@ -31,7 +31,6 @@ from openhands.core.main import create_runtime, run_controller
 from openhands.events.action import CmdRunAction, MessageAction
 from openhands.events.observation import CmdOutputObservation
 from openhands.runtime.base import Runtime
-from openhands.utils.async_utils import call_async_from_sync
 
 AGENT_CLS_TO_FAKE_USER_RESPONSE_FN = {
     'CodeActAgent': functools.partial(
@@ -275,7 +274,6 @@ def process_instance(
     instruction += AGENT_CLS_TO_INST_SUFFIX[metadata.agent_class]
 
     runtime = create_runtime(config)
-    call_async_from_sync(runtime.connect)
     initialize_runtime(runtime, instance)
 
     # Here's how you can run the agent (similar to the `main` function) and get the final task state

evaluation/benchmarks/bird/run_infer.py

@@ -34,7 +34,6 @@ from openhands.core.main import create_runtime, run_controller
 from openhands.events.action import CmdRunAction, MessageAction
 from openhands.events.observation import CmdOutputObservation
 from openhands.runtime.base import Runtime
-from openhands.utils.async_utils import call_async_from_sync
 
 
 def codeact_user_response(state: State) -> str:
@@ -400,7 +399,6 @@ def process_instance(
     instruction += AGENT_CLS_TO_INST_SUFFIX[metadata.agent_class]
 
     runtime = create_runtime(config)
-    call_async_from_sync(runtime.connect)
     initialize_runtime(runtime, instance)
 
     # Here's how you can run the agent (similar to the `main` function) and get the final task state

evaluation/benchmarks/browsing_delegation/run_infer.py

@@ -25,7 +25,6 @@ from openhands.core.config import (
 from openhands.core.logger import openhands_logger as logger
 from openhands.core.main import create_runtime, run_controller
 from openhands.events.action import MessageAction
-from openhands.utils.async_utils import call_async_from_sync
 
 # Only CodeActAgent can delegate to BrowsingAgent
 SUPPORTED_AGENT_CLS = {'CodeActAgent'}
@@ -75,7 +74,6 @@ def process_instance(
     )
 
     runtime = create_runtime(config)
-    call_async_from_sync(runtime.connect)
 
     state: State | None = asyncio.run(
         run_controller(

evaluation/benchmarks/commit0_bench/run_infer.py

@@ -35,7 +35,6 @@ from openhands.events.action import CmdRunAction, MessageAction
 from openhands.events.observation import CmdOutputObservation, ErrorObservation
 from openhands.events.serialization.event import event_to_dict
 from openhands.runtime.base import Runtime
-from openhands.utils.async_utils import call_async_from_sync
 from openhands.utils.shutdown_listener import sleep_if_should_continue
 
 USE_HINT_TEXT = os.environ.get('USE_HINT_TEXT', 'false').lower() == 'true'
@@ -395,7 +394,6 @@ def process_instance(
         logger.info(f'Starting evaluation for instance {instance.instance_id}.')
 
     runtime = create_runtime(config)
-    call_async_from_sync(runtime.connect)
     try:
         initialize_runtime(runtime, instance)
 

evaluation/benchmarks/discoverybench/run_infer.py

@@ -34,7 +34,6 @@ from openhands.core.main import create_runtime, run_controller
 from openhands.events.action import AgentFinishAction, CmdRunAction, MessageAction
 from openhands.events.observation import CmdOutputObservation
 from openhands.runtime.base import Runtime
-from openhands.utils.async_utils import call_async_from_sync
 
 EVALUATION_LLM = 'gpt-4-1106-preview'
 
@@ -282,7 +281,6 @@ def process_instance(
 
     # Here's how you can run the agent (similar to the `main` function) and get the final task state
     runtime = create_runtime(config)
-    call_async_from_sync(runtime.connect)
     initialize_runtime(runtime, instance.data_files)
 
     state: State | None = asyncio.run(

evaluation/benchmarks/gaia/run_infer.py

@@ -31,7 +31,6 @@ from openhands.core.main import create_runtime, run_controller
 from openhands.events.action import AgentFinishAction, CmdRunAction, MessageAction
 from openhands.events.observation import CmdOutputObservation
 from openhands.runtime.base import Runtime
-from openhands.utils.async_utils import call_async_from_sync
 
 DATASET_CACHE_DIR = os.path.join(os.path.dirname(__file__), 'data')
 
@@ -149,7 +148,6 @@ def process_instance(
     logger.info(f'Instruction:\n{instruction}', extra={'msg_type': 'OBSERVATION'})
 
     runtime = create_runtime(config)
-    call_async_from_sync(runtime.connect)
     initialize_runtime(runtime, instance)
 
     # Here's how you can run the agent (similar to the `main` function) and get the final task state

evaluation/benchmarks/gorilla/run_infer.py

@@ -26,7 +26,6 @@ from openhands.core.config import (
 from openhands.core.logger import openhands_logger as logger
 from openhands.core.main import create_runtime, run_controller
 from openhands.events.action import MessageAction
-from openhands.utils.async_utils import call_async_from_sync
 
 AGENT_CLS_TO_FAKE_USER_RESPONSE_FN = {
     'CodeActAgent': codeact_user_response,
@@ -83,7 +82,6 @@ def process_instance(
 
     # Here's how you can run the agent (similar to the `main` function) and get the final task state
     runtime = create_runtime(config)
-    call_async_from_sync(runtime.connect)
     state: State | None = asyncio.run(
         run_controller(
             config=config,

evaluation/benchmarks/gpqa/run_infer.py

@@ -49,7 +49,6 @@ from openhands.events.action import (
     MessageAction,
 )
 from openhands.events.observation import Observation
-from openhands.utils.async_utils import call_async_from_sync
 
 ACTION_FORMAT = """
 <<FINAL_ANSWER||
@@ -215,7 +214,6 @@ Ok now its time to start solving the question. Good luck!
 """
 
     runtime = create_runtime(config)
-    call_async_from_sync(runtime.connect)
     state: State | None = asyncio.run(
         run_controller(
             config=config,

evaluation/benchmarks/humanevalfix/run_infer.py

@@ -39,7 +39,6 @@ from openhands.core.main import create_runtime, run_controller
 from openhands.events.action import CmdRunAction, MessageAction
 from openhands.events.observation import CmdOutputObservation
 from openhands.runtime.base import Runtime
-from openhands.utils.async_utils import call_async_from_sync
 
 IMPORT_HELPER = {
     'python': [
@@ -233,7 +232,6 @@ def process_instance(
 
     # Here's how you can run the agent (similar to the `main` function) and get the final task state
     runtime = create_runtime(config)
-    call_async_from_sync(runtime.connect)
     initialize_runtime(runtime, instance)
     state: State | None = asyncio.run(
         run_controller(

evaluation/benchmarks/logic_reasoning/run_infer.py

@@ -31,7 +31,6 @@ from openhands.events.action import (
 )
 from openhands.events.observation import CmdOutputObservation
 from openhands.runtime.base import Runtime
-from openhands.utils.async_utils import call_async_from_sync
 
 AGENT_CLS_TO_FAKE_USER_RESPONSE_FN = {
     'CodeActAgent': codeact_user_response,
@@ -207,7 +206,6 @@ def process_instance(
     instruction += AGENT_CLS_TO_INST_SUFFIX[metadata.agent_class]
 
     runtime = create_runtime(config)
-    call_async_from_sync(runtime.connect)
     initialize_runtime(runtime, instance)
 
     # Here's how you can run the agent (similar to the `main` function) and get the final task state

evaluation/benchmarks/miniwob/run_infer.py

@@ -41,7 +41,6 @@ from openhands.runtime.browser.browser_env import (
     BROWSER_EVAL_GET_GOAL_ACTION,
     BROWSER_EVAL_GET_REWARDS_ACTION,
 )
-from openhands.utils.async_utils import call_async_from_sync
 
 SUPPORTED_AGENT_CLS = {'BrowsingAgent', 'CodeActAgent'}
 
@@ -146,7 +145,6 @@ def process_instance(
         logger.info(f'Starting evaluation for instance {env_id}.')
 
     runtime = create_runtime(config)
-    call_async_from_sync(runtime.connect)
     task_str, obs = initialize_runtime(runtime)
 
     task_str += (

evaluation/benchmarks/mint/run_infer.py

@@ -35,7 +35,6 @@ from openhands.events.action import (
 )
 from openhands.events.observation import CmdOutputObservation
 from openhands.runtime.base import Runtime
-from openhands.utils.async_utils import call_async_from_sync
 
 
 def codeact_user_response_mint(state: State, task: Task, task_config: dict[str, int]):
@@ -185,7 +184,6 @@ def process_instance(
     )
 
     runtime = create_runtime(config)
-    call_async_from_sync(runtime.connect)
     initialize_runtime(runtime)
 
     state: State | None = asyncio.run(

evaluation/benchmarks/ml_bench/run_infer.py

@@ -43,7 +43,6 @@ from openhands.core.main import create_runtime, run_controller
 from openhands.events.action import CmdRunAction, MessageAction
 from openhands.events.observation import CmdOutputObservation
 from openhands.runtime.base import Runtime
-from openhands.utils.async_utils import call_async_from_sync
 
 config = load_app_config()
 
@@ -235,7 +234,6 @@ def process_instance(instance: Any, metadata: EvalMetadata, reset_logger: bool =
     instruction += AGENT_CLS_TO_INST_SUFFIX[metadata.agent_class]
 
     runtime = create_runtime(config)
-    call_async_from_sync(runtime.connect)
     initialize_runtime(runtime, instance)
 
     # Run the agent

evaluation/benchmarks/scienceagentbench/run_infer.py

@@ -29,7 +29,6 @@ from openhands.core.main import create_runtime, run_controller
 from openhands.events.action import CmdRunAction, MessageAction
 from openhands.events.observation import CmdOutputObservation
 from openhands.runtime.base import Runtime
-from openhands.utils.async_utils import call_async_from_sync
 
 AGENT_CLS_TO_FAKE_USER_RESPONSE_FN = {
     'CodeActAgent': codeact_user_response,
@@ -196,7 +195,6 @@ If the program uses some packages that are incompatible, please figure out alter
 """
 
     runtime = create_runtime(config)
-    call_async_from_sync(runtime.connect)
     initialize_runtime(runtime, instance)
 
     # Here's how you can run the agent (similar to the `main` function) and get the final task state

evaluation/benchmarks/swe_bench/eval_infer.py

@@ -288,7 +288,7 @@ def process_instance(
                                     'model_patch': model_patch,
                                     'instance_id': instance_id,
                                 },
-                                test_log_path=test_output_path,
+                                log_path=test_output_path,
                                 include_tests_status=True,
                             )
                             report = _report[instance_id]

evaluation/benchmarks/swe_bench/run_infer.py

@@ -40,7 +40,6 @@ from openhands.events.action import CmdRunAction, MessageAction
 from openhands.events.observation import CmdOutputObservation, ErrorObservation
 from openhands.events.serialization.event import event_to_dict
 from openhands.runtime.base import Runtime
-from openhands.utils.async_utils import call_async_from_sync
 from openhands.utils.shutdown_listener import sleep_if_should_continue
 
 USE_HINT_TEXT = os.environ.get('USE_HINT_TEXT', 'false').lower() == 'true'
@@ -465,7 +464,6 @@ def process_instance(
             f'This is the {runtime_failure_count + 1}th attempt for instance {instance.instance_id}, setting resource factor to {config.sandbox.remote_runtime_resource_factor}'
         )
     runtime = create_runtime(config)
-    call_async_from_sync(runtime.connect)
 
     try:
         initialize_runtime(runtime, instance)

evaluation/benchmarks/the_agent_company/browsing.py

@@ -7,7 +7,7 @@ import os
 import re
 from dataclasses import dataclass
 from enum import Enum, auto
-from typing import Dict, List, Union
+from typing import Dict, List, Optional, Union
 
 from openhands.core.logger import openhands_logger as logger
 from openhands.events.action import BrowseInteractiveAction
@@ -133,7 +133,7 @@ def parse_content_to_elements(content: str) -> Dict[str, str]:
     return elements
 
 
-def find_matching_anchor(content: str, selector: str) -> str | None:
+def find_matching_anchor(content: str, selector: str) -> Optional[str]:
     """Find the anchor ID that matches the given selector description"""
     elements = parse_content_to_elements(content)
 

evaluation/benchmarks/the_agent_company/run_infer.py

@@ -28,7 +28,6 @@ from openhands.core.main import create_runtime, run_controller
 from openhands.events.action import CmdRunAction, MessageAction
 from openhands.events.observation import BrowserOutputObservation, CmdOutputObservation
 from openhands.runtime.base import Runtime
-from openhands.utils.async_utils import call_async_from_sync
 
 
 def get_config(
@@ -276,7 +275,7 @@ if __name__ == '__main__':
         args.task_image_name, task_short_name, temp_dir, agent_llm_config, agent_config
     )
     runtime: Runtime = create_runtime(config)
-    call_async_from_sync(runtime.connect)
+
     init_task_env(runtime, args.server_hostname, env_llm_config)
 
     dependencies = load_dependencies(runtime)

evaluation/benchmarks/toolqa/run_infer.py

@@ -27,7 +27,6 @@ from openhands.core.main import create_runtime, run_controller
 from openhands.events.action import CmdRunAction, MessageAction
 from openhands.events.observation import CmdOutputObservation
 from openhands.runtime.base import Runtime
-from openhands.utils.async_utils import call_async_from_sync
 
 AGENT_CLS_TO_FAKE_USER_RESPONSE_FN = {
     'CodeActAgent': codeact_user_response,
@@ -105,7 +104,6 @@ def process_instance(instance: Any, metadata: EvalMetadata, reset_logger: bool =
     logger.info(f'Instruction:\n{instruction}', extra={'msg_type': 'OBSERVATION'})
 
     runtime = create_runtime(config)
-    call_async_from_sync(runtime.connect)
     initialize_runtime(runtime)
 
     # Here's how you can run the agent (similar to the `main` function) and get the final task state

evaluation/benchmarks/visualwebarena/run_infer.py

@@ -37,7 +37,6 @@ from openhands.runtime.browser.browser_env import (
     BROWSER_EVAL_GET_GOAL_ACTION,
     BROWSER_EVAL_GET_REWARDS_ACTION,
 )
-from openhands.utils.async_utils import call_async_from_sync
 
 SUPPORTED_AGENT_CLS = {'VisualBrowsingAgent'}
 AGENT_CLS_TO_FAKE_USER_RESPONSE_FN = {
@@ -160,8 +159,6 @@ def process_instance(
         logger.info(f'Starting evaluation for instance {env_id}.')
 
     runtime = create_runtime(config)
-    call_async_from_sync(runtime.connect)
-
     task_str, goal_image_urls = initialize_runtime(runtime)
     initial_user_action = MessageAction(content=task_str, image_urls=goal_image_urls)
     state: State | None = asyncio.run(

evaluation/benchmarks/webarena/run_infer.py

@@ -36,7 +36,6 @@ from openhands.runtime.browser.browser_env import (
     BROWSER_EVAL_GET_GOAL_ACTION,
     BROWSER_EVAL_GET_REWARDS_ACTION,
 )
-from openhands.utils.async_utils import call_async_from_sync
 
 SUPPORTED_AGENT_CLS = {'BrowsingAgent'}
 
@@ -145,7 +144,6 @@ def process_instance(
         logger.info(f'Starting evaluation for instance {env_id}.')
 
     runtime = create_runtime(config)
-    call_async_from_sync(runtime.connect)
     task_str = initialize_runtime(runtime)
 
     state: State | None = asyncio.run(

evaluation/integration_tests/run_infer.py

@@ -30,7 +30,6 @@ from openhands.core.main import create_runtime, run_controller
 from openhands.events.action import MessageAction
 from openhands.events.serialization.event import event_to_dict
 from openhands.runtime.base import Runtime
-from openhands.utils.async_utils import call_async_from_sync
 
 FAKE_RESPONSES = {
     'CodeActAgent': fake_user_response,
@@ -109,7 +108,6 @@ def process_instance(
     # create sandbox and run the agent
     # =============================================
     runtime: Runtime = create_runtime(config)
-    call_async_from_sync(runtime.connect)
     try:
         test_class.initialize_runtime(runtime)
 

frontend/__tests__/components/shared/modals/settings/runtime-size-selector.test.tsx

@@ -0,0 +1,35 @@
+import { screen } from "@testing-library/react";
+import { describe, it, expect } from "vitest";
+import { renderWithProviders } from "test-utils";
+import { RuntimeSizeSelector } from "#/components/shared/modals/settings/runtime-size-selector";
+
+const renderRuntimeSizeSelector = () =>
+  renderWithProviders(<RuntimeSizeSelector isDisabled={false} />);
+
+describe("RuntimeSizeSelector", () => {
+  it("should show both runtime size options", () => {
+    renderRuntimeSizeSelector();
+    // The options are in the hidden select element
+    const select = screen.getByRole("combobox", { hidden: true });
+    expect(select).toHaveValue("1");
+    expect(select).toHaveDisplayValue("1x (2 core, 8G)");
+    expect(select.children).toHaveLength(3); // Empty option + 2 size options
+  });
+
+  it("should show the full description text for disabled options", async () => {
+    renderRuntimeSizeSelector();
+
+    // Click the button to open the dropdown
+    const button = screen.getByRole("button", {
+      name: "1x (2 core, 8G) SETTINGS_FORM$RUNTIME_SIZE_LABEL",
+    });
+    button.click();
+
+    // Wait for the dropdown to open and find the description text
+    const description = await screen.findByText(
+      "Runtime sizes over 1 are disabled by default, please contact contact@all-hands.dev to get access to larger runtimes.",
+    );
+    expect(description).toBeInTheDocument();
+    expect(description).toHaveClass("whitespace-normal", "break-words");
+  });
+});

frontend/package.json

@@ -7,47 +7,47 @@
     "node": ">=20.0.0"
   },
   "dependencies": {
-    "@heroui/react": "2.7.4",
+    "@heroui/react": "2.6.14",
     "@monaco-editor/react": "^4.7.0-rc.0",
-    "@react-router/node": "^7.2.0",
-    "@react-router/serve": "^7.2.0",
+    "@react-router/node": "^7.1.5",
+    "@react-router/serve": "^7.1.5",
     "@react-types/shared": "^3.27.0",
-    "@reduxjs/toolkit": "^2.6.0",
+    "@reduxjs/toolkit": "^2.5.1",
     "@stripe/react-stripe-js": "^3.1.1",
-    "@stripe/stripe-js": "^5.7.0",
-    "@tanstack/react-query": "^5.66.11",
+    "@stripe/stripe-js": "^5.5.0",
+    "@tanstack/react-query": "^5.66.7",
     "@vitejs/plugin-react": "^4.3.2",
     "@xterm/addon-fit": "^0.10.0",
     "@xterm/xterm": "^5.4.0",
-    "axios": "^1.8.1",
+    "axios": "^1.7.9",
     "clsx": "^2.1.1",
     "eslint-config-airbnb-typescript": "^18.0.0",
-    "framer-motion": "^12.4.7",
+    "framer-motion": "^12.4.4",
     "i18next": "^24.2.2",
-    "i18next-browser-languagedetector": "^8.0.4",
+    "i18next-browser-languagedetector": "^8.0.3",
     "i18next-http-backend": "^3.0.2",
-    "isbot": "^5.1.23",
-    "jose": "^6.0.8",
+    "isbot": "^5.1.22",
+    "jose": "^5.10.0",
     "monaco-editor": "^0.52.2",
-    "posthog-js": "^1.225.1",
+    "posthog-js": "^1.219.3",
     "react": "^19.0.0",
     "react-dom": "^19.0.0",
     "react-highlight": "^0.15.0",
     "react-hot-toast": "^2.5.1",
     "react-i18next": "^15.4.1",
-    "react-icons": "^5.5.0",
-    "react-markdown": "^10.0.1",
+    "react-icons": "^5.4.0",
+    "react-markdown": "^9.0.3",
     "react-redux": "^9.2.0",
-    "react-router": "^7.2.0",
+    "react-router": "^7.1.5",
     "react-syntax-highlighter": "^15.6.1",
     "react-textarea-autosize": "^8.5.7",
     "remark-gfm": "^4.0.1",
     "sirv-cli": "^3.0.1",
     "socket.io-client": "^4.8.1",
-    "tailwind-merge": "^3.0.2",
-    "vite": "^6.2.0",
+    "tailwind-merge": "^3.0.1",
+    "vite": "^6.1.0",
     "web-vitals": "^3.5.2",
-    "ws": "^8.18.1"
+    "ws": "^8.18.0"
   },
   "scripts": {
     "dev": "npm run make-i18n && cross-env VITE_MOCK_API=false react-router dev",
@@ -81,14 +81,14 @@
   "devDependencies": {
     "@mswjs/socket.io-binding": "^0.1.1",
     "@playwright/test": "^1.50.1",
-    "@react-router/dev": "^7.2.0",
+    "@react-router/dev": "^7.1.5",
     "@tailwindcss/typography": "^0.5.16",
     "@tanstack/eslint-plugin-query": "^5.66.1",
     "@testing-library/dom": "^10.4.0",
     "@testing-library/jest-dom": "^6.6.1",
     "@testing-library/react": "^16.2.0",
     "@testing-library/user-event": "^14.6.1",
-    "@types/node": "^22.13.8",
+    "@types/node": "^22.13.4",
     "@types/react": "^19.0.8",
     "@types/react-dom": "^19.0.3",
     "@types/react-highlight": "^0.12.8",
@@ -96,13 +96,13 @@
     "@types/ws": "^8.5.14",
     "@typescript-eslint/eslint-plugin": "^7.18.0",
     "@typescript-eslint/parser": "^7.18.0",
-    "@vitest/coverage-v8": "^3.0.7",
+    "@vitest/coverage-v8": "^3.0.6",
     "autoprefixer": "^10.4.20",
     "cross-env": "^7.0.3",
     "eslint": "^8.57.0",
     "eslint-config-airbnb": "^19.0.4",
     "eslint-config-airbnb-typescript": "^18.0.0",
-    "eslint-config-prettier": "^10.0.2",
+    "eslint-config-prettier": "^10.0.1",
     "eslint-plugin-import": "^2.29.1",
     "eslint-plugin-jsx-a11y": "^6.10.2",
     "eslint-plugin-prettier": "^5.2.3",
@@ -113,10 +113,10 @@
     "lint-staged": "^15.4.3",
     "msw": "^2.6.6",
     "postcss": "^8.5.2",
-    "prettier": "^3.5.3",
-    "stripe": "^17.7.0",
+    "prettier": "^3.5.1",
+    "stripe": "^17.5.0",
     "tailwindcss": "^3.4.17",
-    "typescript": "^5.8.2",
+    "typescript": "^5.7.3",
     "vite-plugin-svgr": "^4.2.0",
     "vite-tsconfig-paths": "^5.1.4",
     "vitest": "^3.0.2"

frontend/src/api/open-hands.ts

@@ -174,7 +174,7 @@ class OpenHands {
     code: string,
   ): Promise<GitHubAccessTokenResponse> {
     const { data } = await openHands.post<GitHubAccessTokenResponse>(
-      "/api/keycloak/callback",
+      "/api/github/callback",
       {
         code,
       },

frontend/src/components/features/chat/chat-interface.tsx

@@ -14,10 +14,7 @@ import { generateAgentStateChangeEvent } from "#/services/agent-state-service";
 import { FeedbackModal } from "../feedback/feedback-modal";
 import { useScrollToBottom } from "#/hooks/use-scroll-to-bottom";
 import { TypingIndicator } from "./typing-indicator";
-import {
-  useWsClient,
-  WsClientProviderStatus,
-} from "#/context/ws-client-provider";
+import { useWsClient } from "#/context/ws-client-provider";
 import { Messages } from "./messages";
 import { ChatSuggestions } from "./chat-suggestions";
 import { ActionSuggestions } from "./action-suggestions";
@@ -37,7 +34,7 @@ function getEntryPoint(
 }
 
 export function ChatInterface() {
-  const { send, isLoadingMessages, status, pendingMessages } = useWsClient();
+  const { send, isLoadingMessages } = useWsClient();
   const dispatch = useDispatch();
   const scrollRef = React.useRef<HTMLDivElement>(null);
   const { scrollDomToBottom, onChatBodyScroll, hitBottom } =
@@ -57,9 +54,6 @@ export function ChatInterface() {
   const params = useParams();
   const { mutate: getTrajectory } = useGetTrajectory();
 
-  const isClientDisconnected = status === WsClientProviderStatus.DISCONNECTED;
-  const hasPendingMessages = pendingMessages.length > 0;
-
   const handleSendMessage = async (content: string, files: File[]) => {
     if (messages.length === 0) {
       posthog.capture("initial_query_submitted", {
@@ -82,16 +76,7 @@ export function ChatInterface() {
     const timestamp = new Date().toISOString();
     const pending = true;
     dispatch(addUserMessage({ content, imageUrls, timestamp, pending }));
-
-    // Create the chat message
-    const chatMessage = createChatMessage(content, imageUrls, timestamp);
-
-    // Send or queue the message depending on connection status
-    send(chatMessage);
-
-    // Set agent state to RUNNING when a message is sent
-    send(generateAgentStateChangeEvent(AgentState.RUNNING));
-
+    send(createChatMessage(content, imageUrls, timestamp));
     setMessageToSend(null);
   };
 
@@ -146,20 +131,8 @@ export function ChatInterface() {
         className="flex flex-col grow overflow-y-auto overflow-x-hidden px-4 pt-4 gap-2"
       >
         {isLoadingMessages && (
-          <div className="flex flex-col items-center gap-2">
+          <div className="flex justify-center">
             <LoadingSpinner size="small" />
-            {isClientDisconnected && (
-              <div className="text-sm text-neutral-400">
-                Waiting for client to become ready...
-                {hasPendingMessages && (
-                  <div className="text-xs text-neutral-500 mt-1">
-                    {pendingMessages.length} message
-                    {pendingMessages.length !== 1 ? "s" : ""} will be sent when
-                    connected
-                  </div>
-                )}
-              </div>
-            )}
           </div>
         )}
 
@@ -206,7 +179,7 @@ export function ChatInterface() {
           onSubmit={handleSendMessage}
           onStop={handleStop}
           isDisabled={
-            // Allow input even when loading, but not during confirmation
+            curAgentState === AgentState.LOADING ||
             curAgentState === AgentState.AWAITING_USER_CONFIRMATION
           }
           mode={curAgentState === AgentState.RUNNING ? "stop" : "submit"}

frontend/src/components/features/chat/chat-message.tsx

@@ -57,19 +57,18 @@ export function ChatMessage({
         onClick={handleCopyToClipboard}
         mode={isCopy ? "copied" : "copy"}
       />
-      <div className="text-sm overflow-auto break-words">
-        <Markdown
-          components={{
-            code,
-            ul,
-            ol,
-            a: anchor,
-          }}
-          remarkPlugins={[remarkGfm]}
-        >
-          {message}
-        </Markdown>
-      </div>
+      <Markdown
+        className="text-sm overflow-auto break-words"
+        components={{
+          code,
+          ul,
+          ol,
+          a: anchor,
+        }}
+        remarkPlugins={[remarkGfm]}
+      >
+        {message}
+      </Markdown>
       {children}
     </article>
   );

frontend/src/components/features/chat/expandable-message.tsx

@@ -123,18 +123,17 @@ export function ExpandableMessage({
           )}
         </div>
         {(!headline || showDetails) && (
-          <div className="text-sm overflow-auto">
-            <Markdown
-              components={{
-                code,
-                ul,
-                ol,
-              }}
-              remarkPlugins={[remarkGfm]}
-            >
-              {details}
-            </Markdown>
-          </div>
+          <Markdown
+            className="text-sm overflow-auto"
+            components={{
+              code,
+              ul,
+              ol,
+            }}
+            remarkPlugins={[remarkGfm]}
+          >
+            {details}
+          </Markdown>
         )}
       </div>
     </div>

frontend/src/components/features/controls/agent-status-bar.tsx

@@ -22,11 +22,10 @@ export function AgentStatusBar() {
   const { t, i18n } = useTranslation();
   const { curAgentState } = useSelector((state: RootState) => state.agent);
   const { curStatusMessage } = useSelector((state: RootState) => state.status);
-  const { status, pendingMessages } = useWsClient();
+  const { status } = useWsClient();
   const { notify } = useNotification();
 
   const [statusMessage, setStatusMessage] = React.useState<string>("");
-  const hasPendingMessages = pendingMessages.length > 0;
 
   const updateStatusMessage = () => {
     let message = curStatusMessage.message || "";
@@ -72,13 +71,7 @@ export function AgentStatusBar() {
 
   React.useEffect(() => {
     if (status === WsClientProviderStatus.DISCONNECTED) {
-      if (hasPendingMessages) {
-        setStatusMessage(
-          `Connecting... (${pendingMessages.length} pending message${pendingMessages.length !== 1 ? "s" : ""})`,
-        );
-      } else {
-        setStatusMessage("Connecting...");
-      }
+      setStatusMessage("Connecting...");
     } else {
       setStatusMessage(AGENT_STATUS_MAP[curAgentState].message);
       if (notificationStates.includes(curAgentState)) {
@@ -94,7 +87,7 @@ export function AgentStatusBar() {
         }
       }
     }
-  }, [curAgentState, status, pendingMessages.length, notify, t]);
+  }, [curAgentState, notify, t]);
 
   return (
     <div className="flex flex-col items-center">

frontend/src/components/features/github/github-repo-selector.tsx

@@ -99,6 +99,7 @@ export function GitHubRepositorySelector({
             <AutocompleteItem
               data-testid="github-repo-item"
               key={repo.id}
+              value={repo.id}
               className="data-[selected=true]:bg-default-100"
               textValue={repo.full_name}
             >
@@ -113,6 +114,7 @@ export function GitHubRepositorySelector({
             <AutocompleteItem
               data-testid="github-repo-item"
               key={repo.id}
+              value={repo.id}
               className="data-[selected=true]:bg-default-100"
               textValue={repo.full_name}
             >

frontend/src/components/shared/form-fieldset.tsx

@@ -0,0 +1,43 @@
+import { Autocomplete, AutocompleteItem } from "@heroui/react";
+
+interface FormFieldsetProps {
+  id: string;
+  label: string;
+  items: { key: string; value: string }[];
+  defaultSelectedKey?: string;
+  isClearable?: boolean;
+}
+
+export function FormFieldset({
+  id,
+  label,
+  items,
+  defaultSelectedKey,
+  isClearable,
+}: FormFieldsetProps) {
+  return (
+    <fieldset className="flex flex-col gap-2">
+      <label htmlFor={id} className="font-[500] text-[#A3A3A3] text-xs">
+        {label}
+      </label>
+      <Autocomplete
+        id={id}
+        name={id}
+        aria-label={label}
+        defaultSelectedKey={defaultSelectedKey}
+        isClearable={isClearable}
+        inputProps={{
+          classNames: {
+            inputWrapper: "bg-[#27272A] rounded-md text-sm px-3 py-[10px]",
+          },
+        }}
+      >
+        {items.map((item) => (
+          <AutocompleteItem key={item.key} value={item.key}>
+            {item.value}
+          </AutocompleteItem>
+        ))}
+      </Autocomplete>
+    </fieldset>
+  );
+}

frontend/src/components/shared/inputs/agent-input.tsx

@@ -0,0 +1,46 @@
+import { Autocomplete, AutocompleteItem } from "@heroui/react";
+import { useTranslation } from "react-i18next";
+import { I18nKey } from "#/i18n/declaration";
+
+interface AgentInputProps {
+  isDisabled: boolean;
+  defaultValue: string;
+  agents: string[];
+}
+
+export function AgentInput({
+  isDisabled,
+  defaultValue,
+  agents,
+}: AgentInputProps) {
+  const { t } = useTranslation();
+
+  return (
+    <fieldset data-testid="agent-selector" className="flex flex-col gap-2">
+      <label htmlFor="agent" className="font-[500] text-[#A3A3A3] text-xs">
+        {t(I18nKey.SETTINGS_FORM$AGENT_LABEL)}
+      </label>
+      <Autocomplete
+        isDisabled={isDisabled}
+        isRequired
+        id="agent"
+        aria-label="Agent"
+        data-testid="agent-input"
+        name="agent"
+        defaultSelectedKey={defaultValue}
+        isClearable={false}
+        inputProps={{
+          classNames: {
+            inputWrapper: "bg-[#27272A] rounded-md text-sm px-3 py-[10px]",
+          },
+        }}
+      >
+        {agents.map((agent) => (
+          <AutocompleteItem key={agent} value={agent}>
+            {agent}
+          </AutocompleteItem>
+        ))}
+      </Autocomplete>
+    </fieldset>
+  );
+}

frontend/src/components/shared/inputs/security-analyzers-input.tsx

@@ -0,0 +1,46 @@
+import { Autocomplete, AutocompleteItem } from "@heroui/react";
+import { useTranslation } from "react-i18next";
+import { I18nKey } from "#/i18n/declaration";
+
+interface SecurityAnalyzerInputProps {
+  isDisabled: boolean;
+  defaultValue: string;
+  securityAnalyzers: string[];
+}
+
+export function SecurityAnalyzerInput({
+  isDisabled,
+  defaultValue,
+  securityAnalyzers,
+}: SecurityAnalyzerInputProps) {
+  const { t } = useTranslation();
+
+  return (
+    <fieldset className="flex flex-col gap-2">
+      <label
+        htmlFor="security-analyzer"
+        className="font-[500] text-[#A3A3A3] text-xs"
+      >
+        {t(I18nKey.SETTINGS_FORM$SECURITY_ANALYZER_LABEL)}
+      </label>
+      <Autocomplete
+        isDisabled={isDisabled}
+        id="security-analyzer"
+        name="security-analyzer"
+        aria-label="Security Analyzer"
+        defaultSelectedKey={defaultValue}
+        inputProps={{
+          classNames: {
+            inputWrapper: "bg-[#27272A] rounded-md text-sm px-3 py-[10px]",
+          },
+        }}
+      >
+        {securityAnalyzers.map((analyzer) => (
+          <AutocompleteItem key={analyzer} value={analyzer}>
+            {analyzer}
+          </AutocompleteItem>
+        ))}
+      </Autocomplete>
+    </fieldset>
+  );
+}

frontend/src/components/shared/modals/settings/model-selector.tsx

@@ -100,6 +100,7 @@ export function ModelSelector({
                 <AutocompleteItem
                   data-testid={`provider-item-${provider}`}
                   key={provider}
+                  value={provider}
                 >
                   {mapProvider(provider)}
                 </AutocompleteItem>
@@ -109,7 +110,7 @@ export function ModelSelector({
             {Object.keys(models)
               .filter((provider) => !VERIFIED_PROVIDERS.includes(provider))
               .map((provider) => (
-                <AutocompleteItem key={provider}>
+                <AutocompleteItem key={provider} value={provider}>
                   {mapProvider(provider)}
                 </AutocompleteItem>
               ))}
@@ -147,7 +148,9 @@ export function ModelSelector({
             {models[selectedProvider || ""]?.models
               .filter((model) => VERIFIED_MODELS.includes(model))
               .map((model) => (
-                <AutocompleteItem key={model}>{model}</AutocompleteItem>
+                <AutocompleteItem key={model} value={model}>
+                  {model}
+                </AutocompleteItem>
               ))}
           </AutocompleteSection>
           <AutocompleteSection title="Others">
@@ -157,6 +160,7 @@ export function ModelSelector({
                 <AutocompleteItem
                   data-testid={`model-item-${model}`}
                   key={model}
+                  value={model}
                 >
                   {model}
                 </AutocompleteItem>

frontend/src/components/shared/modals/settings/runtime-size-selector.tsx

@@ -0,0 +1,57 @@
+import { useTranslation } from "react-i18next";
+import { Select, SelectItem } from "@heroui/react";
+import { I18nKey } from "#/i18n/declaration";
+
+interface RuntimeSizeSelectorProps {
+  isDisabled: boolean;
+  defaultValue?: number;
+}
+
+export function RuntimeSizeSelector({
+  isDisabled,
+  defaultValue,
+}: RuntimeSizeSelectorProps) {
+  const { t } = useTranslation();
+
+  return (
+    <fieldset className="flex flex-col gap-2">
+      <label
+        htmlFor="runtime-size"
+        className="font-[500] text-[#A3A3A3] text-xs"
+      >
+        {t(I18nKey.SETTINGS_FORM$RUNTIME_SIZE_LABEL)}
+      </label>
+      <Select
+        data-testid="runtime-size"
+        id="runtime-size"
+        name="runtime-size"
+        defaultSelectedKeys={[String(defaultValue || 1)]}
+        selectedKeys={[String(defaultValue || 1)]}
+        isDisabled={isDisabled}
+        selectionMode="single"
+        disallowEmptySelection
+        aria-label={t(I18nKey.SETTINGS_FORM$RUNTIME_SIZE_LABEL)}
+        classNames={{
+          trigger: "bg-[#27272A] rounded-md text-sm px-3 py-[10px]",
+        }}
+      >
+        <SelectItem key="1" value={1}>
+          1x (2 core, 8G)
+        </SelectItem>
+        <SelectItem
+          key="2"
+          value={2}
+          isDisabled
+          classNames={{
+            description:
+              "whitespace-normal break-words min-w-[300px] max-w-[300px]",
+            base: "min-w-[300px] max-w-[300px]",
+          }}
+          description="Runtime sizes over 1 are disabled by default, please contact contact@all-hands.dev to get access to larger runtimes."
+        >
+          2x (4 core, 16G)
+        </SelectItem>
+      </Select>
+    </fieldset>
+  );
+}

frontend/src/context/ws-client-provider.tsx

@@ -39,14 +39,6 @@ const isMessageAction = (
 ): event is UserMessageAction | AssistantMessageAction =>
   isUserMessage(event) || isAssistantMessage(event);
 
-// Check if an event is an agent state changed observation
-const isAgentStateEvent = (event: Record<string, unknown>): boolean => 
-  isOpenHandsEvent(event) && 
-  "type" in event && 
-  event.type === "observation" && 
-  "observation_id" in event && 
-  event.observation_id === "agent_state_changed";
-
 export enum WsClientProviderStatus {
   CONNECTED,
   DISCONNECTED,
@@ -57,21 +49,15 @@ interface UseWsClient {
   isLoadingMessages: boolean;
   events: Record<string, unknown>[];
   send: (event: Record<string, unknown>) => void;
-  queueMessage: (event: Record<string, unknown>) => void;
-  pendingMessages: Record<string, unknown>[];
 }
 
 const WsClientContext = React.createContext<UseWsClient>({
   status: WsClientProviderStatus.DISCONNECTED,
   isLoadingMessages: true,
   events: [],
-  pendingMessages: [],
   send: () => {
     throw new Error("not connected");
   },
-  queueMessage: () => {
-    throw new Error("not connected");
-  },
 });
 
 interface WsClientProviderProps {
@@ -123,50 +109,26 @@ export function WsClientProvider({
     WsClientProviderStatus.DISCONNECTED,
   );
   const [events, setEvents] = React.useState<Record<string, unknown>[]>([]);
-  const [pendingMessages, setPendingMessages] = React.useState<
-    Record<string, unknown>[]
-  >([]);
-  const [backendReady, setBackendReady] = React.useState(false);
   const lastEventRef = React.useRef<Record<string, unknown> | null>(null);
 
   const messageRateHandler = useRate({ threshold: 250 });
 
-  function queueMessage(event: Record<string, unknown>) {
-    setPendingMessages((prev) => [...prev, event]);
-  }
-
   function send(event: Record<string, unknown>) {
     if (!sioRef.current) {
       EventLogger.error("WebSocket is not connected.");
-      queueMessage(event);
       return;
     }
-    
-    if (!backendReady) {
-      // If backend is not ready yet, queue the message
-      EventLogger.info("Backend not ready, queueing message");
-      queueMessage(event);
-      return;
-    }
-    
     sioRef.current.emit("oh_action", event);
   }
 
   function handleConnect() {
     setStatus(WsClientProviderStatus.CONNECTED);
-    // Don't send queued messages yet - wait for backend ready signal
   }
 
   function handleMessage(event: Record<string, unknown>) {
     if (isOpenHandsEvent(event) && isMessageAction(event)) {
       messageRateHandler.record(new Date().getTime());
     }
-    
-    // Check if this is a state change event indicating backend is ready
-    if (isAgentStateEvent(event)) {
-      setBackendReady(true);
-    }
-    
     setEvents((prevEvents) => [...prevEvents, event]);
     if (!Number.isNaN(parseInt(event.id as string, 10))) {
       lastEventRef.current = event;
@@ -177,7 +139,6 @@ export function WsClientProvider({
 
   function handleDisconnect(data: unknown) {
     setStatus(WsClientProviderStatus.DISCONNECTED);
-    setBackendReady(false);
     const sio = sioRef.current;
     if (!sio) {
       return;
@@ -189,34 +150,11 @@ export function WsClientProvider({
 
   function handleError(data: unknown) {
     setStatus(WsClientProviderStatus.DISCONNECTED);
-    setBackendReady(false);
     updateStatusWhenErrorMessagePresent(data);
   }
 
-  // Watch for backend ready state and send queued messages when ready
-  React.useEffect(() => {
-    if (backendReady && pendingMessages.length > 0 && sioRef.current) {
-      // Backend is ready and we have pending messages
-      EventLogger.info(`Sending ${pendingMessages.length} queued messages`);
-      
-      pendingMessages.forEach((event) => {
-        sioRef.current?.emit("oh_action", event);
-      });
-
-      // Also set the agent state to RUNNING if needed
-      const agentStateEvent = {
-        action: "change_agent_state",
-        args: { agent_state: "running" },
-      };
-      sioRef.current.emit("oh_action", agentStateEvent);
-
-      setPendingMessages([]);
-    }
-  }, [backendReady, pendingMessages.length]);
-
   React.useEffect(() => {
     lastEventRef.current = null;
-    setBackendReady(false);
   }, [conversationId]);
 
   React.useEffect(() => {
@@ -272,11 +210,9 @@ export function WsClientProvider({
       status,
       isLoadingMessages: messageRateHandler.isUnderThreshold,
       events,
-      pendingMessages,
       send,
-      queueMessage,
     }),
-    [status, messageRateHandler.isUnderThreshold, events, pendingMessages],
+    [status, messageRateHandler.isUnderThreshold, events],
   );
 
   return <WsClientContext value={value}>{children}</WsClientContext>;

frontend/src/i18n/declaration.ts

@@ -257,7 +257,6 @@ export enum I18nKey {
   STATUS$ERROR_LLM_AUTHENTICATION = "STATUS$ERROR_LLM_AUTHENTICATION",
   STATUS$ERROR_LLM_SERVICE_UNAVAILABLE = "STATUS$ERROR_LLM_SERVICE_UNAVAILABLE",
   STATUS$ERROR_LLM_INTERNAL_SERVER_ERROR = "STATUS$ERROR_LLM_INTERNAL_SERVER_ERROR",
-  STATUS$ERROR_LLM_OUT_OF_CREDITS = "STATUS$ERROR_LLM_OUT_OF_CREDITS",
   STATUS$ERROR_RUNTIME_DISCONNECTED = "STATUS$ERROR_RUNTIME_DISCONNECTED",
   STATUS$LLM_RETRY = "STATUS$LLM_RETRY",
   AGENT_ERROR$BAD_ACTION = "AGENT_ERROR$BAD_ACTION",
@@ -280,7 +279,6 @@ export enum I18nKey {
   ACTION_MESSAGE$EDIT = "ACTION_MESSAGE$EDIT",
   ACTION_MESSAGE$WRITE = "ACTION_MESSAGE$WRITE",
   ACTION_MESSAGE$BROWSE = "ACTION_MESSAGE$BROWSE",
-  ACTION_MESSAGE$THINK = "ACTION_MESSAGE$THINK",
   OBSERVATION_MESSAGE$RUN = "OBSERVATION_MESSAGE$RUN",
   OBSERVATION_MESSAGE$RUN_IPYTHON = "OBSERVATION_MESSAGE$RUN_IPYTHON",
   OBSERVATION_MESSAGE$READ = "OBSERVATION_MESSAGE$READ",
@@ -308,8 +306,11 @@ export enum I18nKey {
   STATUS$WAITING_FOR_CLIENT = "STATUS$WAITING_FOR_CLIENT",
   SUGGESTIONS$WHAT_TO_BUILD = "SUGGESTIONS$WHAT_TO_BUILD",
   SETTINGS_FORM$ENABLE_DEFAULT_CONDENSER_SWITCH_LABEL = "SETTINGS_FORM$ENABLE_DEFAULT_CONDENSER_SWITCH_LABEL",
+  BUTTON$ENABLE_SOUND = "BUTTON$ENABLE_SOUND",
+  BUTTON$DISABLE_SOUND = "BUTTON$DISABLE_SOUND",
   BUTTON$MARK_HELPFUL = "BUTTON$MARK_HELPFUL",
   BUTTON$MARK_NOT_HELPFUL = "BUTTON$MARK_NOT_HELPFUL",
+  NOTIFICATION$SOUND_ENABLED = "NOTIFICATION$SOUND_ENABLED",
+  NOTIFICATION$SOUND_DISABLED = "NOTIFICATION$SOUND_DISABLED",
   BUTTON$EXPORT_CONVERSATION = "BUTTON$EXPORT_CONVERSATION",
-  BILLING$CLICK_TO_TOP_UP = "BILLING$CLICK_TO_TOP_UP",
 }

frontend/src/services/actions.ts

@@ -57,28 +57,6 @@ const messageActions = {
       store.dispatch(appendJupyterInput(message.args.code));
     }
   },
-  [ActionType.FINISH]: (message: ActionMessage) => {
-    store.dispatch(addAssistantMessage(message.args.final_thought));
-    let successPrediction = "";
-    if (message.args.task_completed === "partial") {
-      successPrediction =
-        "The agent thinks that the task was **completed partially**.";
-    } else if (message.args.task_completed === "false") {
-      successPrediction =
-        "The agent thinks that the task was **not completed**.";
-    } else if (message.args.task_completed === "true") {
-      successPrediction =
-        "The agent thinks that the task was **completed successfully**.";
-    }
-    if (successPrediction) {
-      // if final_thought is not empty, add a new line before the success prediction
-      if (message.args.final_thought) {
-        store.dispatch(addAssistantMessage(`\n${successPrediction}`));
-      } else {
-        store.dispatch(addAssistantMessage(successPrediction));
-      }
-    }
-  },
 };
 
 export function handleActionMessage(message: ActionMessage) {

frontend/src/services/observations.ts

@@ -89,16 +89,6 @@ export function handleObservationMessage(message: ObservationMessage) {
         );
         break;
       case "read":
-        store.dispatch(
-          addAssistantObservation({
-            ...baseObservation,
-            observation,
-            extras: {
-              path: String(message.extras.path || ""),
-            },
-          }),
-        );
-        break;
       case "edit":
         store.dispatch(
           addAssistantObservation({
@@ -106,7 +96,6 @@ export function handleObservationMessage(message: ObservationMessage) {
             observation,
             extras: {
               path: String(message.extras.path || ""),
-              diff: String(message.extras.diff || ""),
             },
           }),
         );

frontend/src/state/chat-slice.ts

@@ -173,14 +173,9 @@ export const chatSlice = createSlice({
           causeMessage.content
         }\n\nOutput:\n\`\`\`\n${content.trim() || "[Command finished execution with no output]"}\n\`\`\``;
         causeMessage.content = content; // Observation content includes the action
-      } else if (observationID === "read") {
-        causeMessage.content = `\`\`\`\n${observation.payload.content}\n\`\`\``; // Content is already truncated by the ACI
-      } else if (observationID === "edit") {
-        if (causeMessage.success) {
-          causeMessage.content = `\`\`\`diff\n${observation.payload.extras.diff}\n\`\`\``; // Content is already truncated by the ACI
-        } else {
-          causeMessage.content = observation.payload.content;
-        }
+      } else if (observationID === "read" || observationID === "edit") {
+        const { content } = observation.payload;
+        causeMessage.content = `\`\`\`${observationID === "edit" ? "diff" : "python"}\n${content}\n\`\`\``; // Content is already truncated by the ACI
       } else if (observationID === "browse") {
         let content = `**URL:** ${observation.payload.extras.url}\n`;
         if (observation.payload.extras.error) {

frontend/src/types/core/actions.ts

@@ -51,8 +51,6 @@ export interface ThinkAction extends OpenHandsActionEvent<"think"> {
 export interface FinishAction extends OpenHandsActionEvent<"finish"> {
   source: "agent";
   args: {
-    final_thought: string;
-    task_completed: "success" | "failure" | "partial";
     outputs: Record<string, unknown>;
     thought: string;
   };

frontend/src/types/core/observations.ts

@@ -70,7 +70,6 @@ export interface EditObservation extends OpenHandsObservationEvent<"edit"> {
   source: "agent";
   extras: {
     path: string;
-    diff: string;
   };
 }
 

frontend/src/utils/event-logger.ts

@@ -37,16 +37,6 @@ class EventLogger {
     }
   }
 
-  /**
-   * Log an info message
-   * @param info The info message
-   */
-  static info(info: string) {
-    if (this.isDevMode) {
-      console.info(info);
-    }
-  }
-
   /**
    * Log an error message
    * @param error The error message

frontend/src/utils/generate-github-auth-url.ts

@@ -5,11 +5,7 @@
  * @returns The URL to redirect to for GitHub OAuth
  */
 export const generateGitHubAuthUrl = (clientId: string, requestUrl: URL) => {
-  const redirectUri = `${requestUrl.origin}/oauth/keycloak/callback`;
-  const authUrl = requestUrl.hostname
-    .replace(/(^|\.)staging\.all-hands\.dev$/, "$1auth.staging.all-hands.dev")
-    .replace(/(^|\.)app\.all-hands\.dev$/, "auth.app.all-hands.dev")
-    .replace(/(^|\.)localhost$/, "auth.staging.all-hands.dev");
-  const scope = "openid email profile offline_access";
-  return `https://${authUrl}/realms/allhands/protocol/openid-connect/auth?client_id=github&response_type=code&redirect_uri=${encodeURIComponent(redirectUri)}&scope=${encodeURIComponent(scope)}`;
+  const redirectUri = `${requestUrl.origin}/oauth/github/callback`;
+  const scope = "repo,user,workflow,offline_access";
+  return `https://github.com/login/oauth/authorize?client_id=${clientId}&redirect_uri=${encodeURIComponent(redirectUri)}&scope=${encodeURIComponent(scope)}`;
 };

microagents/knowledge/docker.md

@@ -1,55 +0,0 @@
----
-name: docker
-type: knowledge
-version: 1.0.0
-agent: CodeActAgent
-triggers:
-- docker
-- container
----
-
-# Docker Installation and Usage Guide
-
-## Installation on Debian/Ubuntu Systems
-
-To install Docker on a Debian/Ubuntu system, follow these steps:
-
-```bash
-# Update package index
-sudo apt-get update
-
-# Install prerequisites
-sudo apt-get install -y apt-transport-https ca-certificates curl gnupg lsb-release
-
-# Add Docker's official GPG key
-curl -fsSL https://download.docker.com/linux/debian/gpg | sudo gpg --dearmor -o /usr/share/keyrings/docker-archive-keyring.gpg
-
-# Set up the stable repository
-echo "deb [arch=amd64 signed-by=/usr/share/keyrings/docker-archive-keyring.gpg] https://download.docker.com/linux/debian $(lsb_release -cs) stable" | sudo tee /etc/apt/sources.list.d/docker.list > /dev/null
-
-# Update package index again
-sudo apt-get update
-
-# Install Docker Engine
-sudo apt-get install -y docker-ce docker-ce-cli containerd.io
-```
-
-## Starting Docker in Container Environments
-
-If you're in a container environment without systemd (like this workspace), start Docker with:
-
-```bash
-# Start Docker daemon in the background
-sudo dockerd > /tmp/docker.log 2>&1 &
-
-# Wait for Docker to initialize
-sleep 5
-```
-
-## Verifying Docker Installation
-
-To verify Docker is working correctly, run the hello-world container:
-
-```bash
-sudo docker run hello-world
-```

microagents/knowledge/kubernetes.md

@@ -1,50 +0,0 @@
----
-name: kubernetes
-type: knowledge
-version: 1.0.0
-agent: CodeActAgent
-triggers:
-- kubernetes
-- k8s
-- kube
----
-
-# Kubernetes Local Development with KIND
-
-## KIND Installation and Setup
-
-KIND (Kubernetes IN Docker) is a tool for running local Kubernetes clusters using Docker containers as nodes. It's designed for testing Kubernetes applications locally.
-
-IMPORTANT: Before you proceed with installation, make sure you have docker installed locally.
-
-### Installation
-
-To install KIND on a Debian/Ubuntu system:
-
-```bash
-# Download KIND binary
-curl -Lo ./kind https://kind.sigs.k8s.io/dl/v0.22.0/kind-linux-amd64
-# Make it executable
-chmod +x ./kind
-# Move to a directory in your PATH
-sudo mv ./kind /usr/local/bin/
-```
-
-To install kubectl:
-
-```bash
-# Download kubectl
-curl -LO "https://dl.k8s.io/release/$(curl -L -s https://dl.k8s.io/release/stable.txt)/bin/linux/amd64/kubectl"
-# Make it executable
-chmod +x kubectl
-# Move to a directory in your PATH
-sudo mv ./kubectl /usr/local/bin/
-```
-
-### Creating a Cluster
-
-Create a basic KIND cluster:
-
-```bash
-kind create cluster
-```

openhands/agenthub/codeact_agent/codeact_agent.py

@@ -2,7 +2,6 @@ import json
 import os
 from collections import deque
 
-import openhands
 import openhands.agenthub.codeact_agent.function_calling as codeact_function_calling
 from openhands.controller.agent import Agent
 from openhands.controller.state.state import State
@@ -77,21 +76,14 @@ class CodeActAgent(Agent):
             f'TOOLS loaded for CodeActAgent: {json.dumps(self.tools, indent=2, ensure_ascii=False).replace("\\n", "\n")}'
         )
         self.prompt_manager = PromptManager(
-            microagent_dir=os.path.join(
-                os.path.dirname(os.path.dirname(openhands.__file__)),
-                'microagents',
-            )
-            if self.config.enable_prompt_extensions
-            else None,
             prompt_dir=os.path.join(os.path.dirname(__file__), 'prompts'),
-            disabled_microagents=self.config.disabled_microagents,
         )
 
         # Create a ConversationMemory instance
         self.conversation_memory = ConversationMemory(self.prompt_manager)
 
         self.condenser = Condenser.from_config(self.config.condenser)
-        logger.debug(f'Using condenser: {type(self.condenser)}')
+        logger.debug(f'Using condenser: {self.condenser}')
 
     def reset(self) -> None:
         """Resets the CodeAct Agent."""
@@ -222,8 +214,6 @@ class CodeActAgent(Agent):
 
             # enhance the user message with additional context based on keywords matched
             if msg.role == 'user':
-                self.prompt_manager.enhance_message(msg)
-
                 # Add double newline between consecutive user messages
                 if prev_role == 'user' and len(msg.content) > 0:
                     # Find the first TextContent in the message to add newlines
@@ -232,7 +222,6 @@ class CodeActAgent(Agent):
                             # If the previous message was also from a user, prepend two newlines to ensure separation
                             content_item.text = '\n\n' + content_item.text
                             break
-
             results.append(msg)
             prev_role = msg.role
 

openhands/agenthub/codeact_agent/function_calling.py

@@ -108,10 +108,7 @@ def response_to_actions(response: ModelResponse) -> list[Action]:
             # AgentFinishAction
             # ================================================
             elif tool_call.function.name == FinishTool['function']['name']:
-                action = AgentFinishAction(
-                    final_thought=arguments.get('message', ''),
-                    task_completed=arguments.get('task_completed', None),
-                )
+                action = AgentFinishAction()
 
             # ================================================
             # LLMBasedFileEditTool (LLM-based file editor, deprecated)

openhands/agenthub/codeact_agent/prompts/microagent_info.j2

@@ -1,8 +0,0 @@
-{% for agent_info in triggered_agents %}
-<EXTRA_INFO>
-The following information has been included based on a keyword match for "{{ agent_info.trigger_word }}".
-It may or may not be relevant to the user's request.
-
-{{ agent_info.agent.content }}
-</EXTRA_INFO>
-{% endfor %}

openhands/agenthub/codeact_agent/tools/finish.py

@@ -1,39 +1,11 @@
 from litellm import ChatCompletionToolParam, ChatCompletionToolParamFunctionChunk
 
-_FINISH_DESCRIPTION = """Signals the completion of the current task or conversation.
-
-Use this tool when:
-- You have successfully completed the user's requested task
-- You cannot proceed further due to technical limitations or missing information
-
-The message should include:
-- A clear summary of actions taken and their results
-- Any next steps for the user
-- Explanation if you're unable to complete the task
-- Any follow-up questions if more information is needed
-
-The task_completed field should be set to True if you believed you have completed the task, and False otherwise.
-"""
+_FINISH_DESCRIPTION = """Finish the interaction when the task is complete OR if the assistant cannot proceed further with the task."""
 
 FinishTool = ChatCompletionToolParam(
     type='function',
     function=ChatCompletionToolParamFunctionChunk(
         name='finish',
         description=_FINISH_DESCRIPTION,
-        parameters={
-            'type': 'object',
-            'required': ['message', 'task_completed'],
-            'properties': {
-                'message': {
-                    'type': 'string',
-                    'description': 'Final message to send to the user',
-                },
-                'task_completed': {
-                    'type': 'string',
-                    'enum': ['true', 'false', 'partial'],
-                    'description': 'Whether you have completed the task.',
-                },
-            },
-        },
     ),
 )

openhands/controller/agent_controller.py

@@ -248,9 +248,8 @@ class AgentController:
             )
             reported = RuntimeError(
                 'There was an unexpected error while running the agent. Please '
-                'report this error to the developers by opening an issue at '
-                'https://github.com/All-Hands-AI/OpenHands. Your session ID is '
-                f' {self.id}. Error type: {e.__class__.__name__}'
+                f'report this error to the developers. Your session ID is {self.id}. '
+                f'Error type: {e.__class__.__name__}'
             )
             if (
                 isinstance(e, litellm.AuthenticationError)
@@ -698,13 +697,10 @@ class AgentController:
             except (ContextWindowExceededError, BadRequestError, OpenAIError) as e:
                 # FIXME: this is a hack until a litellm fix is confirmed
                 # Check if this is a nested context window error
-                # We have to rely on string-matching because LiteLLM doesn't consistently
-                # wrap the failure in a ContextWindowExceededError
                 error_str = str(e).lower()
                 if (
                     'contextwindowexceedederror' in error_str
                     or 'prompt is too long' in error_str
-                    or 'input length and `max_tokens` exceed context limit' in error_str
                     or isinstance(e, ContextWindowExceededError)
                 ):
                     if self.agent.config.enable_history_truncation:

openhands/core/cli.py

@@ -14,12 +14,7 @@ from openhands.core.config import (
 from openhands.core.logger import openhands_logger as logger
 from openhands.core.loop import run_agent_until_done
 from openhands.core.schema import AgentState
-from openhands.core.setup import (
-    create_agent,
-    create_controller,
-    create_runtime,
-    initialize_repository_for_runtime,
-)
+from openhands.core.setup import create_agent, create_controller, create_runtime
 from openhands.events import EventSource, EventStreamSubscriber
 from openhands.events.action import (
     Action,
@@ -114,6 +109,7 @@ async def main(loop: asyncio.AbstractEventLoop):
         sid=sid,
         headless_mode=True,
         agent=agent,
+        selected_repository=config.sandbox.selected_repo,
     )
 
     controller, _ = create_controller(agent, runtime, config)
@@ -169,14 +165,6 @@ async def main(loop: asyncio.AbstractEventLoop):
 
     await runtime.connect()
 
-    # Initialize repository if needed
-    if config.sandbox.selected_repo:
-        initialize_repository_for_runtime(
-            runtime,
-            agent=agent,
-            selected_repository=config.sandbox.selected_repo,
-        )
-
     if initial_user_action:
         # If there's an initial user action, enqueue it and do not prompt again
         event_stream.add_event(initial_user_action, EventSource.USER)

openhands/core/config/app_config.py

@@ -46,6 +46,7 @@ class AppConfig(BaseModel):
         file_uploads_allowed_extensions: Allowed file extensions. `['.*']` allows all.
         cli_multiline_input: Whether to enable multiline input in CLI. When disabled,
             input is read line by line. When enabled, input continues until /exit command.
+        microagents_dir: Directory containing global microagents.
     """
 
     llms: dict[str, LLMConfig] = Field(default_factory=dict)
@@ -82,7 +83,10 @@ class AppConfig(BaseModel):
     daytona_target: str = Field(default='us')
     cli_multiline_input: bool = Field(default=False)
     conversation_max_age_seconds: int = Field(default=864000)  # 10 days in seconds
-    enable_default_condenser: bool = Field(default=True)
+    microagents_dir: str = Field(
+        default='microagents',
+        description='Directory containing global microagents',
+    )
 
     defaults_dict: ClassVar[dict] = {}
 

openhands/core/config/condenser_config.py

@@ -1,8 +1,7 @@
-from typing import Literal, cast
+from typing import Literal
 
-from pydantic import BaseModel, Field, ValidationError
+from pydantic import BaseModel, Field
 
-from openhands.core import logger
 from openhands.core.config.llm_config import LLMConfig
 
 
@@ -11,21 +10,17 @@ class NoOpCondenserConfig(BaseModel):
 
     type: Literal['noop'] = Field('noop')
 
-    model_config = {'extra': 'forbid'}
-
 
 class ObservationMaskingCondenserConfig(BaseModel):
     """Configuration for ObservationMaskingCondenser."""
 
     type: Literal['observation_masking'] = Field('observation_masking')
     attention_window: int = Field(
-        default=100,
+        default=10,
         description='The number of most-recent events where observations will not be masked.',
         ge=1,
     )
 
-    model_config = {'extra': 'forbid'}
-
 
 class RecentEventsCondenserConfig(BaseModel):
     """Configuration for RecentEventsCondenser."""
@@ -39,11 +34,9 @@ class RecentEventsCondenserConfig(BaseModel):
         ge=0,
     )
     max_events: int = Field(
-        default=100, description='Maximum number of events to keep.', ge=1
+        default=10, description='Maximum number of events to keep.', ge=1
     )
 
-    model_config = {'extra': 'forbid'}
-
 
 class LLMSummarizingCondenserConfig(BaseModel):
     """Configuration for LLMCondenser."""
@@ -56,17 +49,13 @@ class LLMSummarizingCondenserConfig(BaseModel):
     # at least one event by default, because the best guess is that it's the user task
     keep_first: int = Field(
         default=1,
-        description='Number of initial events to always keep in history.',
+        description='The number of initial events to condense.',
         ge=0,
     )
     max_size: int = Field(
-        default=100,
-        description='Maximum size of the condensed history before triggering forgetting.',
-        ge=2,
+        default=10, description='Maximum number of events to keep.', ge=1
     )
 
-    model_config = {'extra': 'forbid'}
-
 
 class AmortizedForgettingCondenserConfig(BaseModel):
     """Configuration for AmortizedForgettingCondenser."""
@@ -85,8 +74,6 @@ class AmortizedForgettingCondenserConfig(BaseModel):
         ge=0,
     )
 
-    model_config = {'extra': 'forbid'}
-
 
 class LLMAttentionCondenserConfig(BaseModel):
     """Configuration for LLMAttentionCondenser."""
@@ -108,10 +95,7 @@ class LLMAttentionCondenserConfig(BaseModel):
         ge=0,
     )
 
-    model_config = {'extra': 'forbid'}
 
-
-# Type alias for convenience
 CondenserConfig = (
     NoOpCondenserConfig
     | ObservationMaskingCondenserConfig
@@ -120,121 +104,3 @@ CondenserConfig = (
     | AmortizedForgettingCondenserConfig
     | LLMAttentionCondenserConfig
 )
-
-
-def condenser_config_from_toml_section(
-    data: dict, llm_configs: dict | None = None
-) -> dict[str, CondenserConfig]:
-    """
-    Create a CondenserConfig instance from a toml dictionary representing the [condenser] section.
-
-    For CondenserConfig, the handling is different since it's a union type. The type of condenser
-    is determined by the 'type' field in the section.
-
-    Example:
-    Parse condenser config like:
-        [condenser]
-        type = "noop"
-
-    For condensers that require an LLM config, you can specify the name of an LLM config:
-        [condenser]
-        type = "llm"
-        llm_config = "my_llm"  # References [llm.my_llm] section
-
-    Args:
-        data: The TOML dictionary representing the [condenser] section.
-        llm_configs: Optional dictionary of LLMConfig objects keyed by name.
-
-    Returns:
-        dict[str, CondenserConfig]: A mapping where the key "condenser" corresponds to the configuration.
-    """
-
-    # Initialize the result mapping
-    condenser_mapping: dict[str, CondenserConfig] = {}
-
-    # Process config
-    try:
-        # Determine which condenser type to use based on 'type' field
-        condenser_type = data.get('type', 'noop')
-
-        # Handle LLM config reference if needed
-        if (
-            condenser_type in ('llm', 'llm_attention')
-            and 'llm_config' in data
-            and isinstance(data['llm_config'], str)
-        ):
-            llm_config_name = data['llm_config']
-            if llm_configs and llm_config_name in llm_configs:
-                # Replace the string reference with the actual LLMConfig object
-                data_copy = data.copy()
-                data_copy['llm_config'] = llm_configs[llm_config_name]
-                config = create_condenser_config(condenser_type, data_copy)
-            else:
-                logger.openhands_logger.warning(
-                    f"LLM config '{llm_config_name}' not found for condenser. Using default LLMConfig."
-                )
-                # Create a default LLMConfig if the referenced one doesn't exist
-                data_copy = data.copy()
-                # Try to use the fallback 'llm' config
-                if llm_configs is not None:
-                    data_copy['llm_config'] = llm_configs.get('llm')
-                config = create_condenser_config(condenser_type, data_copy)
-        else:
-            config = create_condenser_config(condenser_type, data)
-
-        condenser_mapping['condenser'] = config
-    except (ValidationError, ValueError) as e:
-        logger.openhands_logger.warning(
-            f'Invalid condenser configuration: {e}. Using NoOpCondenserConfig.'
-        )
-        # Default to NoOpCondenserConfig if config fails
-        config = NoOpCondenserConfig()
-        condenser_mapping['condenser'] = config
-
-    return condenser_mapping
-
-
-# For backward compatibility
-from_toml_section = condenser_config_from_toml_section
-
-
-def create_condenser_config(condenser_type: str, data: dict) -> CondenserConfig:
-    """
-    Create a CondenserConfig instance based on the specified type.
-
-    Args:
-        condenser_type: The type of condenser to create.
-        data: The configuration data.
-
-    Returns:
-        A CondenserConfig instance.
-
-    Raises:
-        ValueError: If the condenser type is unknown.
-        ValidationError: If the provided data fails validation for the condenser type.
-    """
-    # Mapping of condenser types to their config classes
-    condenser_classes = {
-        'noop': NoOpCondenserConfig,
-        'observation_masking': ObservationMaskingCondenserConfig,
-        'recent': RecentEventsCondenserConfig,
-        'llm': LLMSummarizingCondenserConfig,
-        'amortized': AmortizedForgettingCondenserConfig,
-        'llm_attention': LLMAttentionCondenserConfig,
-    }
-
-    if condenser_type not in condenser_classes:
-        raise ValueError(f'Unknown condenser type: {condenser_type}')
-
-    # Create and validate the config using direct instantiation
-    # Explicitly handle ValidationError to provide more context
-    try:
-        config_class = condenser_classes[condenser_type]
-        # Use type casting to help mypy understand the return type
-        return cast(CondenserConfig, config_class(**data))
-    except ValidationError as e:
-        # Just re-raise with a more descriptive message, but don't try to pass the errors
-        # which can cause compatibility issues with different pydantic versions
-        raise ValueError(
-            f"Validation failed for condenser type '{condenser_type}': {e}"
-        )

openhands/core/config/llm_config.py

@@ -48,7 +48,7 @@ class LLMConfig(BaseModel):
         reasoning_effort: The effort to put into reasoning. This is a string that can be one of 'low', 'medium', 'high', or 'none'. Exclusive for o1 models.
     """
 
-    model: str = Field(default='claude-3-7-sonnet-20250219')
+    model: str = Field(default='claude-3-5-sonnet-20241022')
     api_key: SecretStr | None = Field(default=None)
     base_url: str | None = Field(default=None)
     api_version: str | None = Field(default=None)

openhands/core/config/sandbox_config.py

@@ -17,7 +17,6 @@ class SandboxConfig(BaseModel):
         remote_runtime_api_timeout: The timeout for the remote runtime API requests.
         enable_auto_lint: Whether to enable auto-lint.
         use_host_network: Whether to use the host network.
-        runtime_binding_address: The binding address for the runtime ports.  It specifies which network interface on the host machine Docker should bind the runtime ports to.
         initialize_plugins: Whether to initialize plugins.
         force_rebuild_runtime: Whether to force rebuild the runtime image.
         runtime_extra_deps: The extra dependencies to install in the runtime image (typically used for evaluation).
@@ -61,7 +60,6 @@ class SandboxConfig(BaseModel):
         default=False
     )  # once enabled, OpenHands would lint files after editing
     use_host_network: bool = Field(default=False)
-    runtime_binding_address: str = Field(default='127.0.0.1')
     runtime_extra_build_args: list[str] | None = Field(default=None)
     initialize_plugins: bool = Field(default=True)
     force_rebuild_runtime: bool = Field(default=False)
@@ -72,7 +70,7 @@ class SandboxConfig(BaseModel):
     close_delay: int = Field(default=15)
     remote_runtime_resource_factor: int = Field(default=1)
     enable_gpu: bool = Field(default=False)
-    docker_runtime_kwargs: dict | None = Field(default=None)
+    docker_runtime_kwargs: str | None = Field(default=None)
     selected_repo: str | None = Field(default=None)
 
     model_config = {'extra': 'forbid'}

openhands/core/config/utils.py

@@ -12,11 +12,9 @@ import toml
 from dotenv import load_dotenv
 from pydantic import BaseModel, SecretStr, ValidationError
 
-from openhands import __version__
 from openhands.core import logger
 from openhands.core.config.agent_config import AgentConfig
 from openhands.core.config.app_config import AppConfig
-from openhands.core.config.condenser_config import condenser_config_from_toml_section
 from openhands.core.config.config_utils import (
     OH_DEFAULT_AGENT,
     OH_MAX_ITERATIONS,
@@ -195,44 +193,6 @@ def load_from_toml(cfg: AppConfig, toml_file: str = 'config.toml') -> None:
             # Re-raise ValueError from SandboxConfig.from_toml_section
             raise ValueError('Error in [sandbox] section in config.toml')
 
-    # Process condenser section if present
-    if 'condenser' in toml_config:
-        try:
-            # Pass the LLM configs to the condenser config parser
-            condenser_mapping = condenser_config_from_toml_section(
-                toml_config['condenser'], cfg.llms
-            )
-            # Assign the default condenser configuration to the default agent configuration
-            if 'condenser' in condenser_mapping:
-                # Get the default agent config and assign the condenser config to it
-                default_agent_config = cfg.get_agent_config()
-                default_agent_config.condenser = condenser_mapping['condenser']
-                logger.openhands_logger.debug(
-                    'Default condenser configuration loaded from config toml and assigned to default agent'
-                )
-        except (TypeError, KeyError, ValidationError) as e:
-            logger.openhands_logger.warning(
-                f'Cannot parse [condenser] config from toml, values have not been applied.\nError: {e}'
-            )
-    # If no condenser section is in toml but enable_default_condenser is True,
-    # set LLMSummarizingCondenserConfig as default
-    elif cfg.enable_default_condenser:
-        from openhands.core.config.condenser_config import LLMSummarizingCondenserConfig
-
-        # Get default agent config
-        default_agent_config = cfg.get_agent_config()
-
-        # Create default LLM summarizing condenser config
-        default_condenser = LLMSummarizingCondenserConfig(
-            llm_config=cfg.get_llm_config(),  # Use default LLM config
-        )
-
-        # Set as default condenser
-        default_agent_config.condenser = default_condenser
-        logger.openhands_logger.debug(
-            'Default LLM summarizing condenser assigned to default agent (no condenser in config)'
-        )
-
     # Process extended section if present
     if 'extended' in toml_config:
         try:
@@ -243,15 +203,7 @@ def load_from_toml(cfg: AppConfig, toml_file: str = 'config.toml') -> None:
             )
 
     # Check for unknown sections
-    known_sections = {
-        'core',
-        'extended',
-        'agent',
-        'llm',
-        'security',
-        'sandbox',
-        'condenser',
-    }
+    known_sections = {'core', 'extended', 'agent', 'llm', 'security', 'sandbox'}
     for key in toml_config:
         if key.lower() not in known_sections:
             logger.openhands_logger.warning(f'Unknown section [{key}] in {toml_file}')
@@ -540,6 +492,8 @@ def parse_arguments() -> argparse.Namespace:
     args = parser.parse_args()
 
     if args.version:
+        from openhands import __version__
+
         print(f'OpenHands version: {__version__}')
         sys.exit(0)
 

openhands/core/logger.py

@@ -6,21 +6,15 @@ import sys
 import traceback
 from datetime import datetime
 from types import TracebackType
-from typing import Any, Literal, Mapping, TextIO
+from typing import Any, Literal, Mapping
 
 import litellm
-from pythonjsonlogger.json import JsonFormatter
 from termcolor import colored
 
 LOG_LEVEL = os.getenv('LOG_LEVEL', 'INFO').upper()
 DEBUG = os.getenv('DEBUG', 'False').lower() in ['true', '1', 'yes']
 DEBUG_LLM = os.getenv('DEBUG_LLM', 'False').lower() in ['true', '1', 'yes']
 
-# Structured logs with JSON, disabled by default
-LOG_JSON = os.getenv('LOG_JSON', 'False').lower() in ['true', '1', 'yes']
-LOG_JSON_LEVEL_KEY = os.getenv('LOG_JSON_LEVEL_KEY', 'level')
-
-
 # Configure litellm logging based on DEBUG_LLM
 if DEBUG_LLM:
     confirmation = input(
@@ -83,14 +77,7 @@ class StackInfoFilter(logging.Filter):
     def filter(self, record: logging.LogRecord) -> bool:
         if record.levelno >= logging.ERROR:
             # LogRecord attributes are dynamically typed
-
-            # Capture the current stack trace as a string
-            stack = traceback.format_stack()
-            # Remove the last entries which are related to the logging machinery
-            stack = stack[:-3]  # Adjust this number if needed
-            # Join the stack frames into a single string
-            stack_str = ''.join(stack)
-            setattr(record, 'stack_info', stack_str)
+            setattr(record, 'stack_info', True)
             setattr(record, 'exc_info', sys.exc_info())
         return True
 
@@ -300,36 +287,10 @@ def get_file_handler(
     file_name = f'openhands_{timestamp}.log'
     file_handler = logging.FileHandler(os.path.join(log_dir, file_name))
     file_handler.setLevel(log_level)
-    if LOG_JSON:
-        file_handler.setFormatter(json_formatter())
-    else:
-        file_handler.setFormatter(file_formatter)
+    file_handler.setFormatter(file_formatter)
     return file_handler
 
 
-def json_formatter():
-    return JsonFormatter(
-        '{message}{levelname}',
-        style='{',
-        rename_fields={'levelname': LOG_JSON_LEVEL_KEY},
-        timestamp=True,
-    )
-
-
-def json_log_handler(
-    level: int = logging.INFO,
-    _out: TextIO = sys.stdout,
-) -> logging.Handler:
-    """
-    Configure logger instance for structured logging as json lines.
-    """
-
-    handler = logging.StreamHandler(_out)
-    handler.setLevel(level)
-    handler.setFormatter(json_formatter())
-    return handler
-
-
 # Set up logging
 logging.basicConfig(level=logging.ERROR)
 
@@ -367,11 +328,7 @@ if current_log_level == logging.DEBUG:
     LOG_TO_FILE = True
     openhands_logger.debug('DEBUG mode enabled.')
 
-if LOG_JSON:
-    openhands_logger.addHandler(json_log_handler(current_log_level))
-else:
-    openhands_logger.addHandler(get_console_handler(current_log_level))
-
+openhands_logger.addHandler(get_console_handler(current_log_level))
 openhands_logger.addFilter(SensitiveDataFilter(openhands_logger.name))
 openhands_logger.propagate = False
 openhands_logger.debug('Logging initialized')

openhands/core/main.py

@@ -18,9 +18,9 @@ from openhands.core.schema import AgentState
 from openhands.core.setup import (
     create_agent,
     create_controller,
+    create_memory,
     create_runtime,
     generate_sid,
-    initialize_repository_for_runtime,
 )
 from openhands.events import EventSource, EventStreamSubscriber
 from openhands.events.action import MessageAction, NullAction
@@ -30,7 +30,6 @@ from openhands.events.observation import AgentStateChangedObservation
 from openhands.events.serialization import event_from_dict
 from openhands.io import read_input, read_task
 from openhands.runtime.base import Runtime
-from openhands.utils.async_utils import call_async_from_sync
 
 
 class FakeUserResponseFunc(Protocol):
@@ -99,20 +98,12 @@ async def run_controller(
             sid=sid,
             headless_mode=headless_mode,
             agent=agent,
+            selected_repository=config.sandbox.selected_repo,
         )
-        # Connect to the runtime
-        call_async_from_sync(runtime.connect)
-
-        # Initialize repository if needed
-        if config.sandbox.selected_repo:
-            initialize_repository_for_runtime(
-                runtime,
-                agent=agent,
-                selected_repository=config.sandbox.selected_repo,
-            )
 
     event_stream = runtime.event_stream
 
+
     replay_events: list[Event] | None = None
     if config.replay_trajectory_path:
         logger.info('Trajectory replay is enabled')

openhands/core/message_format.md

@@ -7,7 +7,6 @@ OpenHands uses its own `Message` class (`openhands/core/message.py`) which provi
 ## Class Structure
 
 Our `Message` class (`openhands/core/message.py`):
-
 ```python
 class Message(BaseModel):
     role: Literal['user', 'system', 'assistant', 'tool']
@@ -23,14 +22,13 @@ class Message(BaseModel):
 ```
 
 litellm's `Message` class (`litellm/types/utils.py`):
-
 ```python
 class Message(OpenAIObject):
-    content: str | None
+    content: Optional[str]
     role: Literal["assistant", "user", "system", "tool", "function"]
-    tool_calls: List[ChatCompletionMessageToolCall] | None
-    function_call: FunctionCall | None
-    audio: ChatCompletionAudioResponse | None = None
+    tool_calls: Optional[List[ChatCompletionMessageToolCall]]
+    function_call: Optional[FunctionCall]
+    audio: Optional[ChatCompletionAudioResponse] = None
 ```
 
 ## How It Works
@@ -38,7 +36,6 @@ class Message(OpenAIObject):
 1. **Message Creation**: Our `Message` class is a Pydantic model that supports rich content (text and images) through its `content` field.
 
 2. **Serialization**: The class uses Pydantic's `@model_serializer` to convert messages into dictionaries that litellm can understand. We have two serialization methods:
-
    ```python
    def _string_serializer(self) -> dict:
        # convert content to a single string
@@ -58,7 +55,6 @@ class Message(OpenAIObject):
    ```
 
    The appropriate serializer is chosen based on the message's capabilities:
-
    ```python
    @model_serializer
    def serialize_model(self) -> dict:
@@ -68,13 +64,11 @@ class Message(OpenAIObject):
    ```
 
 3. **Tool Call Handling**: Tool calls require special attention in serialization because:
-
    - They need to work with litellm's API calls (which accept both dicts and objects)
    - They need to be properly serialized for token counting
    - They need to maintain compatibility with different LLM providers' formats
 
 4. **litellm Integration**: When we pass our messages to `litellm.completion()`, litellm doesn't care about the message class type - it works with the dictionary representation. This works because:
-
    - litellm's transformation code (e.g., `litellm/llms/anthropic/chat/transformation.py`) processes messages based on their structure, not their type
    - our serialization produces dictionaries that match litellm's expected format
    - litellm handles rich content by looking at the message structure, supporting both simple string content and lists of content items
@@ -84,7 +78,6 @@ class Message(OpenAIObject):
 ### Token Counting
 
 To use litellm's token counter, we need to make sure that all message components (including tool calls) are properly serialized to dictionaries. This is because:
-
 - litellm's token counter expects dictionary structures
 - Tool calls need to be included in the token count
 - Different providers may count tokens differently for structured content

openhands/core/schema/action.py

@@ -82,5 +82,8 @@ class ActionTypeSchema(BaseModel):
     SEND_PR: str = Field(default='send_pr')
     """Send a PR to github."""
 
+    RECALL: str = Field(default='recall')
+    """Retrieves data from a file or other storage."""
+
 
 ActionType = ActionTypeSchema()

openhands/core/schema/observation.py

@@ -49,5 +49,8 @@ class ObservationTypeSchema(BaseModel):
     CONDENSE: str = Field(default='condense')
     """Result of a condensation operation."""
 
+    RECALL: str = Field(default='recall')
+    """Result of a recall operation."""
+
 
 ObservationType = ObservationTypeSchema()

openhands/core/setup.py

@@ -16,11 +16,13 @@ from openhands.core.logger import openhands_logger as logger
 from openhands.events import EventStream
 from openhands.events.event import Event
 from openhands.llm.llm import LLM
+from openhands.memory.memory import Memory
 from openhands.microagent.microagent import BaseMicroAgent
 from openhands.runtime import get_runtime_cls
 from openhands.runtime.base import Runtime
 from openhands.security import SecurityAnalyzer, options
 from openhands.storage import get_file_store
+from openhands.utils.async_utils import call_async_from_sync
 
 
 def create_runtime(
@@ -28,19 +30,18 @@ def create_runtime(
     sid: str | None = None,
     headless_mode: bool = True,
     agent: Agent | None = None,
+    selected_repository: str | None = None,
+    github_token: SecretStr | None = None,
 ) -> Runtime:
     """Create a runtime for the agent to run on.
 
-    Args:
-        config: The app config.
-        sid: (optional) The session id. IMPORTANT: please don't set this unless you know what you're doing.
-            Set it to incompatible value will cause unexpected behavior on RemoteRuntime.
-        headless_mode: Whether the agent is run in headless mode. `create_runtime` is typically called within evaluation scripts,
-            where we don't want to have the VSCode UI open, so it defaults to True.
-        agent: (optional) The agent instance to use for configuring the runtime.
-
-    Returns:
-        The created Runtime instance (not yet connected or initialized).
+    config: The app config.
+    sid: (optional) The session id. IMPORTANT: please don't set this unless you know what you're doing.
+        Set it to incompatible value will cause unexpected behavior on RemoteRuntime.
+    headless_mode: Whether the agent is run in headless mode. `create_runtime` is typically called within evaluation scripts,
+        where we don't want to have the VSCode UI open, so it defaults to True.
+    selected_repository: (optional) The GitHub repository to use.
+    github_token: (optional) The GitHub token to use.
     """
     # if sid is provided on the command line, use it as the name of the event stream
     # otherwise generate it on the basis of the configured jwt_secret
@@ -74,30 +75,8 @@ def create_runtime(
         headless_mode=headless_mode,
     )
 
-    logger.debug(
-        f'Runtime created with plugins: {[plugin.name for plugin in runtime.plugins]}'
-    )
+    call_async_from_sync(runtime.connect)
 
-    return runtime
-
-
-def initialize_repository_for_runtime(
-    runtime: Runtime,
-    agent: Agent | None = None,
-    selected_repository: str | None = None,
-    github_token: SecretStr | None = None,
-) -> str | None:
-    """Initialize the repository for the runtime.
-
-    Args:
-        runtime: The runtime to initialize the repository for.
-        agent: (optional) The agent to load microagents for.
-        selected_repository: (optional) The GitHub repository to use.
-        github_token: (optional) The GitHub token to use.
-
-    Returns:
-        The repository directory path if a repository was cloned, None otherwise.
-    """
     # clone selected repository if provided
     repo_directory = None
     github_token = (
@@ -120,7 +99,44 @@ def initialize_repository_for_runtime(
         agent.prompt_manager.load_microagents(microagents)
         agent.prompt_manager.set_repository_info(selected_repository, repo_directory)
 
-    return repo_directory
+    logger.debug(
+        f'Runtime initialized with plugins: {[plugin.name for plugin in runtime.plugins]}'
+    )
+
+    return runtime
+
+
+def create_memory(
+    microagents_dir: str,
+    agent: Agent,
+    runtime: Runtime,
+    event_stream: EventStream,
+    selected_repository: str | None = None,
+) -> Memory:
+    # If the agent config has disabled microagents, use them
+    disabled_microagents = agent.config.disabled_microagents
+
+    mem = Memory(
+        event_stream=event_stream,
+        microagents_dir=microagents_dir,
+        disabled_microagents=disabled_microagents,
+    )
+
+    if agent.prompt_manager and runtime:
+        # sets available hosts
+        mem.set_runtime_info(runtime.web_hosts)
+
+        # loads microagents from repo/.openhands/microagents
+        microagents: list[BaseMicroAgent] = runtime.get_microagents_from_selected_repo(
+            selected_repository
+        )
+        mem.load_user_workspace_microagents(microagents)
+
+        if selected_repository:
+            repo_directory = selected_repository.split('/')[1]
+            if repo_directory:
+                mem.set_repository_info(selected_repository, repo_directory)
+    return mem
 
 
 def create_agent(config: AppConfig) -> Agent:
@@ -132,6 +148,7 @@ def create_agent(config: AppConfig) -> Agent:
         config=agent_config,
     )
 
+
     return agent
 
 

openhands/events/action/agent.py

@@ -1,5 +1,4 @@
 from dataclasses import dataclass, field
-from enum import Enum
 from typing import Any
 
 from openhands.core.schema import ActionType
@@ -34,26 +33,16 @@ class AgentSummarizeAction(Action):
         return ret
 
 
-class AgentFinishTaskCompleted(Enum):
-    FALSE = 'false'
-    PARTIAL = 'partial'
-    TRUE = 'true'
-
-
 @dataclass
 class AgentFinishAction(Action):
     """An action where the agent finishes the task.
 
     Attributes:
-        final_thought (str): The message to send to the user.
-        task_completed (enum): Whether the agent believes the task has been completed.
-        outputs (dict): The other outputs of the agent, for instance "content".
+        outputs (dict): The outputs of the agent, for instance "content".
         thought (str): The agent's explanation of its actions.
         action (str): The action type, namely ActionType.FINISH.
     """
 
-    final_thought: str = ''
-    task_completed: AgentFinishTaskCompleted | None = None
     outputs: dict[str, Any] = field(default_factory=dict)
     thought: str = ''
     action: str = ActionType.FINISH
@@ -106,3 +95,15 @@ class AgentDelegateAction(Action):
     @property
     def message(self) -> str:
         return f"I'm asking {self.agent} for help with this task."
+
+
+@dataclass
+class RecallAction(Action):
+    # This action is used for retrieving data, e.g., from memory or a knowledge base.
+    query: dict[str, Any] = field(default_factory=dict)
+    thought: str = ''
+    action: str = ActionType.RECALL
+
+    @property
+    def message(self) -> str:
+        return f'Retrieved data for: {self.query}'

openhands/events/observation/agent.py

@@ -27,6 +27,13 @@ class AgentCondensationObservation(Observation):
         return self.content
 
 
+@dataclass
+class RecallObservation(Observation):
+    """The output of a recall action."""
+
+    observation: str = ObservationType.RECALL
+
+
 @dataclass
 class AgentThinkObservation(Observation):
     """The output of a think action.

openhands/events/observation/files.py

@@ -62,12 +62,7 @@ class FileEditObservation(Observation):
     new_content: str | None = None
     observation: str = ObservationType.EDIT
     impl_source: FileEditSource = FileEditSource.LLM_BASED_EDIT
-    diff: str | None = (
-        None  # The raw diff between old and new content, used in OH_ACI mode
-    )
-    _diff_cache: str | None = (
-        None  # Cache for the diff visualization, used in LLM-based editing mode
-    )
+    _diff_cache: str | None = None  # Cache for the diff visualization
 
     @property
     def message(self) -> str:
@@ -131,7 +126,7 @@ class FileEditObservation(Observation):
         n_context_lines: int = 2,
         change_applied: bool = True,
     ) -> str:
-        """Visualize the diff of the file edit. Used in the LLM-based editing mode.
+        """Visualize the diff of the file edit.
 
         Instead of showing the diff line by line, this function shows each hunk
         of changes as a separate entity.

openhands/events/serialization/action.py

@@ -8,6 +8,7 @@ from openhands.events.action.agent import (
     AgentRejectAction,
     AgentThinkAction,
     ChangeAgentStateAction,
+    RecallAction,
 )
 from openhands.events.action.browse import BrowseInteractiveAction, BrowseURLAction
 from openhands.events.action.commands import (
@@ -37,6 +38,7 @@ actions = (
     AgentDelegateAction,
     ChangeAgentStateAction,
     MessageAction,
+    RecallAction,
 )
 
 ACTION_TYPE_TO_CLASS = {action_class.action: action_class for action_class in actions}  # type: ignore[attr-defined]

openhands/events/stream.py

@@ -27,6 +27,7 @@ class EventStreamSubscriber(str, Enum):
     RESOLVER = 'openhands_resolver'
     SERVER = 'server'
     RUNTIME = 'runtime'
+    MEMORY = 'memory'
     MAIN = 'main'
     TEST = 'test'
 

openhands/integrations/github/github_service.py

@@ -21,12 +21,7 @@ class GitHubService:
     token: SecretStr = SecretStr('')
     refresh = False
 
-    def __init__(
-        self,
-        user_id: str | None = None,
-        idp_token: SecretStr | None = None,
-        token: SecretStr | None = None,
-    ):
+    def __init__(self, user_id: str | None = None, token: SecretStr | None = None):
         self.user_id = user_id
 
         if token:
@@ -51,9 +46,6 @@ class GitHubService:
     async def get_latest_token(self) -> SecretStr:
         return self.token
 
-    async def get_latest_provider_token(self) -> SecretStr:
-        return self.token
-
     async def _fetch_data(
         self, url: str, params: dict | None = None
     ) -> tuple[Any, dict]:

openhands/llm/llm.py

@@ -76,8 +76,6 @@ REASONING_EFFORT_SUPPORTED_MODELS = [
 MODELS_WITHOUT_STOP_WORDS = [
     'o1-mini',
     'o1-preview',
-    'o1',
-    'o1-2024-12-17',
 ]
 
 
@@ -219,8 +217,9 @@ class LLM(RetryMixin, DebugMixin):
                     kwargs['stop'] = STOP_WORDS
 
                 mock_fncall_tools = kwargs.pop('tools')
-                # tool_choice should not be specified when mocking function calling
-                kwargs.pop('tool_choice', None)
+                kwargs['tool_choice'] = (
+                    'none'  # force no tool calling because we're mocking it - without it, it will cause issue with sglang
+                )
 
             # if we have no messages, something went very wrong
             if not messages:

openhands/memory/condenser/__init__.py

@@ -1,4 +1,4 @@
 import openhands.memory.condenser.impl  # noqa F401 (we import this to get the condensers registered)
 from openhands.memory.condenser.condenser import Condenser, get_condensation_metadata
 
-__all__ = ['Condenser', 'get_condensation_metadata', 'CONDENSER_REGISTRY']
+__all__ = ['Condenser', 'get_condensation_metadata']

openhands/memory/memory.py

@@ -0,0 +1,202 @@
+from openhands.core.logger import openhands_logger as logger
+from openhands.events.action.agent import RecallAction
+from openhands.events.action.message import MessageAction
+from openhands.events.event import Event, EventSource
+from openhands.events.observation.agent import (
+    RecallObservation,
+)
+from openhands.events.stream import EventStream, EventStreamSubscriber
+from openhands.microagent import (
+    BaseMicroAgent,
+    KnowledgeMicroAgent,
+    RepoMicroAgent,
+    load_microagents_from_dir,
+)
+from openhands.utils.prompt import PromptManager, RepositoryInfo, RuntimeInfo
+
+
+class Memory:
+    """
+    Memory is a component that listens to the EventStream for either user MessageAction (to create
+    a RecallAction) or a RecallAction (to produce a RecallObservation).
+    """
+
+    def __init__(
+        self,
+        event_stream: EventStream,
+        microagents_dir: str,
+        disabled_microagents: list[str] | None = None,
+    ):
+        self.event_stream = event_stream
+        self.microagents_dir = microagents_dir
+        self.disabled_microagents = disabled_microagents or []
+        # Subscribe to events
+        self.event_stream.subscribe(
+            EventStreamSubscriber.MEMORY,
+            self.on_event,
+            'Memory',
+        )
+        # Load global microagents (Knowledge + Repo).
+        self._load_global_microagents()
+
+        # Additional placeholders to store user workspace microagents if needed
+        self.repo_microagents: dict[str, RepoMicroAgent] = {}
+        self.knowledge_microagents: dict[str, KnowledgeMicroAgent] = {}
+
+        # Track whether we've seen the first user message
+        self._first_user_message_seen = False
+
+        # Store repository / runtime info to send them to the templating later
+        self.repository_info: RepositoryInfo | None = None
+        self.runtime_info: RuntimeInfo | None = None
+
+        # TODO: enable_prompt_extensions
+
+    def _load_global_microagents(self) -> None:
+        """
+        Loads microagents from the global microagents_dir.
+        This is effectively what used to happen in PromptManager.
+        """
+        repo_agents, knowledge_agents, _ = load_microagents_from_dir(
+            self.microagents_dir
+        )
+        for name, agent in knowledge_agents.items():
+            if name in self.disabled_microagents:
+                continue
+            if isinstance(agent, KnowledgeMicroAgent):
+                self.knowledge_microagents[name] = agent
+        for name, agent in repo_agents.items():
+            if name in self.disabled_microagents:
+                continue
+            if isinstance(agent, RepoMicroAgent):
+                self.repo_microagents[name] = agent
+
+    def set_repository_info(self, repo_name: str, repo_directory: str) -> None:
+        """Store repository info so we can reference it in an observation."""
+        self.repository_info = RepositoryInfo(repo_name, repo_directory)
+        self.prompt_manager.set_repository_info(self.repository_info)
+
+    def set_runtime_info(self, runtime_hosts: dict[str, int]) -> None:
+        """Store runtime info (web hosts, ports, etc.)."""
+        # e.g. { '127.0.0.1': 8080 }
+        self.runtime_info = RuntimeInfo(available_hosts=runtime_hosts)
+        self.prompt_manager.set_runtime_info(self.runtime_info)
+
+    def on_event(self, event: Event):
+        """Handle an event from the event stream."""
+        if isinstance(event, MessageAction):
+            if event.source == 'user':
+                # If this is the first user message, create and add a RecallObservation
+                # with info about repo and runtime.
+                if not self._first_user_message_seen:
+                    self._first_user_message_seen = True
+                    self._on_first_user_message(event)
+                    # continue with the next handler, to include microagents if suitable for this user message
+            self._on_user_message_action(event)
+        elif isinstance(event, RecallAction):
+            self._on_recall_action(event)
+
+    def _on_first_user_message(self, event: MessageAction):
+        """Create and add to the stream a RecallObservation carrying info about repo and runtime."""
+        # Build the same text that used to be appended to the first user message
+        repo_instructions = ''
+        assert (
+            len(self.repo_microagents) <= 1
+        ), f'Expecting at most one repo microagent, but found {len(self.repo_microagents)}: {self.repo_microagents.keys()}'
+        for microagent in self.repo_microagents.values():
+            # We assume these are the repo instructions
+            if repo_instructions:
+                repo_instructions += '\n\n'
+            repo_instructions += microagent.content
+
+        # Now wrap it in a RecallObservation, rather than altering the user message:
+        obs = RecallObservation(
+            content=self.prompt_manager.build_additional_info_text(repo_instructions)
+        )
+        self.event_stream.add_event(obs, EventSource.ENVIRONMENT)
+
+    def _on_user_message_action(self, event: MessageAction):
+        """Replicates old microagent logic: if a microagent triggers on user text,
+        we embed it in an <extra_info> block and post a RecallObservation."""
+        if event.source != 'user':
+            return
+
+        # If there's no text, do nothing
+        user_text = event.content.strip()
+        if not user_text:
+            return
+        # Gather all triggered microagents
+        microagent_blocks = []
+        for name, agent in self.knowledge_microagents.items():
+            trigger = agent.match_trigger(user_text)
+            if trigger:
+                logger.info("Microagent '%s' triggered by keyword '%s'", name, trigger)
+                micro_text = (
+                    f'<extra_info>\n'
+                    f'The following information has been included based on a keyword match for "{trigger}". '
+                    f"It may or may not be relevant to the user's request.\n\n"
+                    f'{agent.content}\n'
+                    f'</extra_info>'
+                )
+                microagent_blocks.append(micro_text)
+
+        if microagent_blocks:
+            # Combine all triggered microagents into a single RecallObservation
+            combined_text = '\n'.join(microagent_blocks)
+            obs = RecallObservation(content=combined_text)
+            self.event_stream.add_event(
+                obs, event.source if event.source else EventSource.ENVIRONMENT
+            )
+
+    def _on_recall_action(self, event: RecallAction):
+        """If a RecallAction explicitly arrives, handle it."""
+        assert isinstance(event, RecallAction)
+
+        user_query = event.query.get('keywords', [])
+        matched_content = self.find_microagent_content(user_query)
+        obs = RecallObservation(content=matched_content)
+        self.event_stream.add_event(
+            obs, event.source if event.source else EventSource.ENVIRONMENT
+        )
+
+    def find_microagent_content(self, keywords: list[str]) -> str:
+        """Replicate the same microagent logic."""
+        matched_texts: list[str] = []
+        for name, agent in self.knowledge_microagents.items():
+            for kw in keywords:
+                trigger = agent.match_trigger(kw)
+                if trigger:
+                    logger.info(
+                        "Microagent '%s' triggered by explicit RecallAction keyword '%s'",
+                        name,
+                        trigger,
+                    )
+                    block = (
+                        f'<extra_info>\n'
+                        f"(via RecallAction) Included knowledge from microagent '{name}', triggered by '{trigger}'\n\n"
+                        f'{agent.content}\n'
+                        f'</extra_info>'
+                    )
+                    matched_texts.append(block)
+        return '\n'.join(matched_texts)
+
+    def load_user_workspace_microagents(
+        self, user_microagents: list[BaseMicroAgent]
+    ) -> None:
+        """
+        If you want to load microagents from a user's cloned repo or workspace directory,
+        call this from agent_session or setup once the workspace is cloned.
+        """
+        logger.info(
+            'Loading user workspace microagents: %s', [m.name for m in user_microagents]
+        )
+        for ma in user_microagents:
+            if ma.name in self.disabled_microagents:
+                continue
+            if isinstance(ma, KnowledgeMicroAgent):
+                self.knowledge_microagents[ma.name] = ma
+            elif isinstance(ma, RepoMicroAgent):
+                self.repo_microagents[ma.name] = ma
+
+    def set_prompt_manager(self, prompt_manager: PromptManager):
+        self.prompt_manager = prompt_manager

openhands/resolver/resolve_issue.py

@@ -202,7 +202,7 @@ async def process_issue(
         timeout=300,
     )
 
-    if os.getenv('GITLAB_CI') == 'true':
+    if os.getenv('GITLAB_CI') == 'True':
         sandbox_config.local_runtime_url = os.getenv(
             'LOCAL_RUNTIME_URL', 'http://localhost'
         )
@@ -651,7 +651,7 @@ def main() -> None:
     if not token:
         raise ValueError('Token is required.')
 
-    platform = identify_token(token, repo)
+    platform = identify_token(token)
     if platform == Platform.INVALID:
         raise ValueError('Token is invalid.')
 

openhands/resolver/utils.py

@@ -22,37 +22,18 @@ class Platform(Enum):
     GITLAB = 2
 
 
-def identify_token(token: str, repo: str | None = None) -> Platform:
+def identify_token(token: str) -> Platform:
     """
     Identifies whether a token belongs to GitHub or GitLab.
 
     Parameters:
         token (str): The personal access token to check.
-        repo (str): Repository in format "owner/repo" for GitHub Actions token validation.
 
     Returns:
         Platform: "GitHub" if the token is valid for GitHub,
              "GitLab" if the token is valid for GitLab,
              "Invalid" if the token is not recognized by either.
     """
-    # Try GitHub Actions token format (Bearer) with repo endpoint if repo is provided
-    if repo:
-        github_repo_url = f'https://api.github.com/repos/{repo}'
-        github_bearer_headers = {
-            'Authorization': f'Bearer {token}',
-            'Accept': 'application/vnd.github+json',
-        }
-
-        try:
-            github_repo_response = requests.get(
-                github_repo_url, headers=github_bearer_headers, timeout=5
-            )
-            if github_repo_response.status_code == 200:
-                return Platform.GITHUB
-        except requests.RequestException as e:
-            print(f'Error connecting to GitHub API (repo check): {e}')
-
-    # Try GitHub PAT format (token)
     github_url = 'https://api.github.com/user'
     github_headers = {'Authorization': f'token {token}'}
 
@@ -63,7 +44,6 @@ def identify_token(token: str, repo: str | None = None) -> Platform:
     except requests.RequestException as e:
         print(f'Error connecting to GitHub API: {e}')
 
-    # Try GitLab token
     gitlab_url = 'https://gitlab.com/api/v4/user'
     gitlab_headers = {'Authorization': f'Bearer {token}'}
 

openhands/runtime/action_execution_server.py

@@ -25,7 +25,6 @@ from fastapi.security import APIKeyHeader
 from openhands_aci.editor.editor import OHEditor
 from openhands_aci.editor.exceptions import ToolError
 from openhands_aci.editor.results import ToolResult
-from openhands_aci.utils.diff import get_diff
 from pydantic import BaseModel
 from starlette.background import BackgroundTask
 from starlette.exceptions import HTTPException as StarletteHTTPException
@@ -90,7 +89,7 @@ def _execute_file_editor(
     new_str: str | None = None,
     insert_line: int | None = None,
     enable_linting: bool = False,
-) -> tuple[str, tuple[str | None, str | None]]:
+) -> str:
     """Execute file editor command and handle exceptions.
 
     Args:
@@ -105,7 +104,7 @@ def _execute_file_editor(
         enable_linting: Whether to enable linting
 
     Returns:
-        tuple: A tuple containing the output string and a tuple of old and new file content
+        str: Result string from the editor operation
     """
     result: ToolResult | None = None
     try:
@@ -123,13 +122,13 @@ def _execute_file_editor(
         result = ToolResult(error=e.message)
 
     if result.error:
-        return f'ERROR:\n{result.error}', (None, None)
+        return f'ERROR:\n{result.error}'
 
     if not result.output:
         logger.warning(f'No output from file_editor for {path}')
-        return '', (None, None)
+        return ''
 
-    return result.output, (result.old_content, result.new_content)
+    return result.output
 
 
 class ActionExecutor:
@@ -317,7 +316,7 @@ class ActionExecutor:
     async def read(self, action: FileReadAction) -> Observation:
         assert self.bash_session is not None
         if action.impl_source == FileReadSource.OH_ACI:
-            result_str, _ = _execute_file_editor(
+            result_str = _execute_file_editor(
                 self.file_editor,
                 command='view',
                 path=action.path,
@@ -434,7 +433,7 @@ class ActionExecutor:
 
     async def edit(self, action: FileEditAction) -> Observation:
         assert action.impl_source == FileEditSource.OH_ACI
-        result_str, (old_content, new_content) = _execute_file_editor(
+        result_str = _execute_file_editor(
             self.file_editor,
             command=action.command,
             path=action.path,
@@ -451,11 +450,6 @@ class ActionExecutor:
             old_content=action.old_str,
             new_content=action.new_str,
             impl_source=FileEditSource.OH_ACI,
-            diff=get_diff(
-                old_contents=old_content or '',
-                new_contents=new_content or '',
-                filepath=action.path,
-            ),
         )
 
     async def browse(self, action: BrowseURLAction) -> Observation:

openhands/runtime/base.py

@@ -222,7 +222,7 @@ class Runtime(FileEditRuntimeMixin):
             if isinstance(event, CmdRunAction):
                 if self.github_user_id and '$GITHUB_TOKEN' in event.command:
                     gh_client = GithubServiceImpl(user_id=self.github_user_id)
-                    token = await gh_client.get_latest_provider_token()
+                    token = await gh_client.get_latest_token()
                     if token:
                         export_cmd = CmdRunAction(
                             f"export GITHUB_TOKEN='{token.get_secret_value()}'"

openhands/runtime/builder/docker.py

@@ -19,30 +19,19 @@ class DockerRuntimeBuilder(RuntimeBuilder):
 
         version_info = self.docker_client.version()
         server_version = version_info.get('Version', '').replace('-', '.')
-        self.is_podman = (
-            version_info.get('Components')[0].get('Name').startswith('Podman')
-        )
-        if (
-            tuple(map(int, server_version.split('.')[:2])) < (18, 9)
-            and not self.is_podman
-        ):
+        if tuple(map(int, server_version.split('.')[:2])) < (18, 9):
             raise AgentRuntimeBuildError(
                 'Docker server version must be >= 18.09 to use BuildKit'
             )
 
-        if self.is_podman and tuple(map(int, server_version.split('.')[:2])) < (4, 9):
-            raise AgentRuntimeBuildError('Podman server version must be >= 4.9.0')
-
         self.rolling_logger = RollingLogger(max_lines=10)
 
     @staticmethod
-    def check_buildx(is_podman: bool = False):
+    def check_buildx():
         """Check if Docker Buildx is available"""
         try:
             result = subprocess.run(
-                ['docker' if not is_podman else 'podman', 'buildx', 'version'],
-                capture_output=True,
-                text=True,
+                ['docker', 'buildx', 'version'], capture_output=True, text=True
             )
             return result.returncode == 0
         except FileNotFoundError:
@@ -79,18 +68,12 @@ class DockerRuntimeBuilder(RuntimeBuilder):
         self.docker_client = docker.from_env()
         version_info = self.docker_client.version()
         server_version = version_info.get('Version', '').split('+')[0].replace('-', '.')
-        self.is_podman = (
-            version_info.get('Components')[0].get('Name').startswith('Podman')
-        )
-        if tuple(map(int, server_version.split('.'))) < (18, 9) and not self.is_podman:
+        if tuple(map(int, server_version.split('.'))) < (18, 9):
             raise AgentRuntimeBuildError(
                 'Docker server version must be >= 18.09 to use BuildKit'
             )
 
-        if self.is_podman and tuple(map(int, server_version.split('.'))) < (4, 9):
-            raise AgentRuntimeBuildError('Podman server version must be >= 4.9.0')
-
-        if not DockerRuntimeBuilder.check_buildx(self.is_podman):
+        if not DockerRuntimeBuilder.check_buildx():
             # when running openhands in a container, there might not be a "docker"
             # binary available, in which case we need to download docker binary.
             # since the official openhands app image is built from debian, we use
@@ -127,7 +110,7 @@ class DockerRuntimeBuilder(RuntimeBuilder):
         target_image_tag = tags[1].split(':')[1] if len(tags) > 1 else None
 
         buildx_cmd = [
-            'docker' if not self.is_podman else 'podman',
+            'docker',
             'buildx',
             'build',
             '--progress=plain',
@@ -156,7 +139,7 @@ class DockerRuntimeBuilder(RuntimeBuilder):
         buildx_cmd.append(path)  # must be last!
 
         self.rolling_logger.start(
-            f'================ {buildx_cmd[0].upper()} BUILD STARTED ================'
+            '================ DOCKER BUILD STARTED ================'
         )
 
         try:

openhands/runtime/impl/daytona/README.md

@@ -2,80 +2,22 @@
 
 [Daytona](https://www.daytona.io/) is a platform that provides a secure and elastic infrastructure for running AI-generated code. It provides all the necessary features for an AI Agent to interact with a codebase. It provides a Daytona SDK with official Python and TypeScript interfaces for interacting with Daytona, enabling you to programmatically manage development environments and execute code.
 
-## Quick Start
+## Getting started
 
-### Step 1: Retrieve Your Daytona API Key
-1. Visit the [Daytona Dashboard](https://app.daytona.io/dashboard/keys).
-2. Click **"Create Key"**.
-3. Enter a name for your key and confirm the creation.
-4. Once the key is generated, copy it.
+1. Sign in at https://app.daytona.io/
+
+1. Generate and copy your API key
+
+1. Set the following environment variables before running the OpenHands app on your local machine or via a `docker run` command:
 
-### Step 2: Set Your API Key as an Environment Variable
-Run the following command in your terminal, replacing `<your-api-key>` with the actual key you copied:
 ```bash
-export DAYTONA_API_KEY="<your-api-key>"
+    RUNTIME="daytona"
+    DAYTONA_API_KEY="<your-api-key>"
 ```
-
-This step ensures that OpenHands can authenticate with the Daytona platform when it runs.
-
-### Step 3: Run OpenHands Locally Using Docker
-To start the latest version of OpenHands on your machine, execute the following command in your terminal:
-```bash
-bash -i <(curl -sL https://get.daytona.io/openhands)
-```
-
-#### What This Command Does:
-- Downloads the latest OpenHands release script.
-- Runs the script in an interactive Bash session.
-- Automatically pulls and runs the OpenHands container using Docker.
-Once executed, OpenHands should be running locally and ready for use.
-
-
-## Manual Initialization
-
-### Step 1: Set the `OPENHANDS_VERSION` Environment Variable
-Run the following command in your terminal, replacing `<openhands-release>` with the latest release's version seen in the [main README.md file](https://github.com/All-Hands-AI/OpenHands?tab=readme-ov-file#-quick-start):
+Optionally, if you don't want your sandboxes to default to the US region, set:
 
 ```bash
-export OPENHANDS_VERSION="<openhands-release>"  # e.g. 0.27
-```
-
-### Step 2: Retrieve Your Daytona API Key
-1. Visit the [Daytona Dashboard](https://app.daytona.io/dashboard/keys).
-2. Click **"Create Key"**.
-3. Enter a name for your key and confirm the creation.
-4. Once the key is generated, copy it.
-
-### Step 3: Set Your API Key as an Environment Variable:
-Run the following command in your terminal, replacing `<your-api-key>` with the actual key you copied:
-```bash
-export DAYTONA_API_KEY="<your-api-key>"
-```
-
-### Step 4: Run the following `docker` command:
-This command pulls and runs the OpenHands container using Docker. Once executed, OpenHands should be running locally and ready for use.
-
-```bash
-docker run -it --rm --pull=always \
-    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:${OPENHANDS_VERSION}-nikolaik \
-    -e LOG_ALL_EVENTS=true \
-    -e RUNTIME=daytona \
-    -e DAYTONA_API_KEY=${DAYTONA_API_KEY} \
-    -v ~/.openhands-state:/.openhands-state \
-    -p 3000:3000 \
-    --name openhands-app \
-    docker.all-hands.dev/all-hands-ai/openhands:${OPENHANDS_VERSION}
-```
-
-> **Tip:** If you don't want your sandboxes to default to the US region, you can set the `DAYTONA_TARGET` environment variable to `eu`
-
-### Running OpenHands Locally Without Docker
-
-Alternatively, if you want to run the OpenHands app on your local machine using `make run` without Docker, make sure to set the following environment variables first:
-
-```bash
-export RUNTIME="daytona"
-export DAYTONA_API_KEY="<your-api-key>"
+    DAYTONA_TARGET="eu"
 ```
 
 ## Documentation

openhands/runtime/impl/docker/docker_runtime.py

@@ -201,29 +201,16 @@ class DockerRuntime(ActionExecutionClient):
         port_mapping: dict[str, list[dict[str, str]]] | None = None
         if not use_host_network:
             port_mapping = {
-                f'{self._container_port}/tcp': [
-                    {
-                        'HostPort': str(self._host_port),
-                        'HostIp': self.config.sandbox.runtime_binding_address,
-                    }
-                ],
+                f'{self._container_port}/tcp': [{'HostPort': str(self._host_port)}],
             }
 
             if self.vscode_enabled:
                 port_mapping[f'{self._vscode_port}/tcp'] = [
-                    {
-                        'HostPort': str(self._vscode_port),
-                        'HostIp': self.config.sandbox.runtime_binding_address,
-                    }
+                    {'HostPort': str(self._vscode_port)}
                 ]
 
             for port in self._app_ports:
-                port_mapping[f'{port}/tcp'] = [
-                    {
-                        'HostPort': str(port),
-                        'HostIp': self.config.sandbox.runtime_binding_address,
-                    }
-                ]
+                port_mapping[f'{port}/tcp'] = [{'HostPort': str(port)}]
         else:
             self.log(
                 'warn',
@@ -319,7 +306,6 @@ class DockerRuntime(ActionExecutionClient):
         self.container = self.docker_client.containers.get(self.container_name)
         if self.container.status == 'exited':
             self.container.start()
-
         config = self.container.attrs['Config']
         for env_var in config['Env']:
             if env_var.startswith('port='):
@@ -327,18 +313,11 @@ class DockerRuntime(ActionExecutionClient):
                 self._container_port = self._host_port
             elif env_var.startswith('VSCODE_PORT='):
                 self._vscode_port = int(env_var.split('VSCODE_PORT=')[1])
-
         self._app_ports = []
-        exposed_ports = config.get('ExposedPorts')
-        if exposed_ports:
-            for exposed_port in exposed_ports.keys():
-                exposed_port = int(exposed_port.split('/tcp')[0])
-                if (
-                    exposed_port != self._host_port
-                    and exposed_port != self._vscode_port
-                ):
-                    self._app_ports.append(exposed_port)
-
+        for exposed_port in config['ExposedPorts'].keys():
+            exposed_port = int(exposed_port.split('/tcp')[0])
+            if exposed_port != self._host_port and exposed_port != self._vscode_port:
+                self._app_ports.append(exposed_port)
         self.api_url = f'{self.config.sandbox.local_runtime_url}:{self._container_port}'
         self.log(
             'debug',

openhands/runtime/impl/e2b/e2b_runtime.py

@@ -1,4 +1,4 @@
-from typing import Callable
+from typing import Callable, Optional
 
 from openhands.core.config import AppConfig
 from openhands.events.action import (
@@ -27,7 +27,7 @@ class E2BRuntime(Runtime):
         sid: str = 'default',
         plugins: list[PluginRequirement] | None = None,
         sandbox: E2BSandbox | None = None,
-        status_callback: Callable | None = None,
+        status_callback: Optional[Callable] = None,
     ):
         super().__init__(
             config,

openhands/runtime/impl/local/local_runtime.py

@@ -7,7 +7,7 @@ import shutil
 import subprocess
 import tempfile
 import threading
-from typing import Callable
+from typing import Callable, Optional
 
 import requests
 import tenacity
@@ -155,7 +155,7 @@ class LocalRuntime(ActionExecutionClient):
 
         self.api_url = f'{self.config.sandbox.local_runtime_url}:{self._host_port}'
         self.status_callback = status_callback
-        self.server_process: subprocess.Popen[str] | None = None
+        self.server_process: Optional[subprocess.Popen[str]] = None
         self.action_semaphore = threading.Semaphore(1)  # Ensure one action at a time
 
         # Update env vars

openhands/runtime/impl/remote/remote_runtime.py

@@ -1,5 +1,5 @@
 import os
-from typing import Callable
+from typing import Callable, Optional
 from urllib.parse import urlparse
 
 import requests
@@ -42,7 +42,7 @@ class RemoteRuntime(ActionExecutionClient):
         sid: str = 'default',
         plugins: list[PluginRequirement] | None = None,
         env_vars: dict[str, str] | None = None,
-        status_callback: Callable | None = None,
+        status_callback: Optional[Callable] = None,
         attach_to_existing: bool = False,
         headless_mode: bool = True,
         github_user_id: str | None = None,

openhands/runtime/utils/vscode-extensions/memory-monitor/README.md

@@ -45,4 +45,4 @@ This extension is part of the OpenHands project. To modify or extend it:
 
 ## License
 
-This extension is licensed under the MIT license.
+This extension is licensed under the MIT license.

openhands/runtime/utils/vscode-extensions/memory-monitor/extension.js

@@ -4,30 +4,30 @@ const MemoryMonitor = require('./memory_monitor');
 function activate(context) {
     // Create memory monitor instance
     const memoryMonitor = new MemoryMonitor();
-
+    
     // Store the context in the memory monitor
     memoryMonitor.context = context;
-
+    
     // Register memory monitor start command
     let startMonitorCommand = vscode.commands.registerCommand('openhands-memory-monitor.startMemoryMonitor', function () {
         memoryMonitor.start();
     });
-
+    
     // Register memory monitor stop command
     let stopMonitorCommand = vscode.commands.registerCommand('openhands-memory-monitor.stopMemoryMonitor', function () {
         memoryMonitor.stop();
     });
-
+    
     // Register memory details command
     let showMemoryDetailsCommand = vscode.commands.registerCommand('openhands-memory-monitor.showMemoryDetails', function () {
         memoryMonitor.showDetails();
     });
-
+    
     // Add all commands to subscriptions
     context.subscriptions.push(startMonitorCommand);
     context.subscriptions.push(stopMonitorCommand);
     context.subscriptions.push(showMemoryDetailsCommand);
-
+    
     // Start memory monitoring by default
     memoryMonitor.start();
 }
@@ -39,4 +39,4 @@ function deactivate() {
 module.exports = {
     activate,
     deactivate
-}
+}

openhands/runtime/utils/vscode-extensions/memory-monitor/memory_monitor.js

@@ -21,15 +21,15 @@ class MemoryMonitor {
 
         this.isMonitoring = true;
         this.statusBarItem.show();
-
+        
         // Initial update
         this.updateMemoryInfo();
-
+        
         // Set interval for updates
         this.intervalId = setInterval(() => {
             this.updateMemoryInfo();
         }, interval);
-
+        
         vscode.window.showInformationMessage('Memory monitoring started');
     }
 
@@ -41,7 +41,7 @@ class MemoryMonitor {
         this.isMonitoring = false;
         clearInterval(this.intervalId);
         this.statusBarItem.hide();
-
+        
         vscode.window.showInformationMessage('Memory monitoring stopped');
     }
 
@@ -49,18 +49,18 @@ class MemoryMonitor {
         const totalMem = os.totalmem();
         const freeMem = os.freemem();
         const usedMem = totalMem - freeMem;
-
+        
         // Calculate memory usage percentage
         const memUsagePercent = Math.round((usedMem / totalMem) * 100);
-
+        
         // Format memory values to MB
         const usedMemMB = Math.round(usedMem / (1024 * 1024));
         const totalMemMB = Math.round(totalMem / (1024 * 1024));
-
+        
         // Update status bar
         this.statusBarItem.text = `$(pulse) Mem: ${memUsagePercent}%`;
         this.statusBarItem.tooltip = `Memory Usage: ${usedMemMB}MB / ${totalMemMB}MB`;
-
+        
         // Store memory data in history
         this.memoryHistory.push({
             timestamp: new Date(),
@@ -69,7 +69,7 @@ class MemoryMonitor {
             memUsagePercent,
             processMemory: process.memoryUsage()
         });
-
+        
         // Limit history length
         if (this.memoryHistory.length > this.maxHistoryLength) {
             this.memoryHistory.shift();
@@ -86,7 +86,7 @@ class MemoryMonitor {
                 enableScripts: true
             }
         );
-
+        
         // Set up message handler for real-time updates
         panel.webview.onDidReceiveMessage(
             message => {
@@ -97,60 +97,60 @@ class MemoryMonitor {
             undefined,
             this.context ? this.context.subscriptions : []
         );
-
+        
         // Initial update
         this.updateWebviewContent(panel);
-
+        
         // Handle panel disposal
         panel.onDidDispose(() => {
             // Clean up any resources if needed
         }, null, this.context ? this.context.subscriptions : []);
     }
-
+    
     updateWebviewContent(panel) {
         // Get system memory info
         const totalMem = os.totalmem();
         const freeMem = os.freemem();
         const usedMem = totalMem - freeMem;
-
+        
         // Format memory values
         const usedMemMB = Math.round(usedMem / (1024 * 1024));
         const freeMemMB = Math.round(freeMem / (1024 * 1024));
         const totalMemMB = Math.round(totalMem / (1024 * 1024));
-
+        
         // Get process memory usage
         const processMemory = process.memoryUsage();
         const rss = Math.round(processMemory.rss / (1024 * 1024));
         const heapTotal = Math.round(processMemory.heapTotal / (1024 * 1024));
         const heapUsed = Math.round(processMemory.heapUsed / (1024 * 1024));
-
+        
         // Get process information
         this.processMonitor.getProcessInfo((error, processInfo) => {
             if (error) {
                 console.error('Error getting process info:', error);
                 return;
             }
-
+            
             // Create HTML content for the webview
             const htmlContent = this.generateHtmlReport(
-                usedMemMB, freeMemMB, totalMemMB,
+                usedMemMB, freeMemMB, totalMemMB, 
                 rss, heapTotal, heapUsed,
                 processInfo
             );
-
+            
             // Set the webview's HTML content
             panel.webview.html = htmlContent;
         });
     }
-
+    
     generateHtmlReport(usedMemMB, freeMemMB, totalMemMB, rss, heapTotal, heapUsed, processInfo) {
         // Create memory usage history data for chart
         const memoryLabels = this.memoryHistory.map((entry, index) => index);
         const memoryData = this.memoryHistory.map(entry => entry.memUsagePercent);
-        const heapData = this.memoryHistory.map(entry =>
+        const heapData = this.memoryHistory.map(entry => 
             Math.round(entry.processMemory.heapUsed / (1024 * 1024))
         );
-
+        
         // Format process info table
         let processTable = '';
         if (processInfo && processInfo.processes) {
@@ -174,7 +174,7 @@ class MemoryMonitor {
                 </table>
             `;
         }
-
+        
         return `
             <!DOCTYPE html>
             <html lang="en">
@@ -237,7 +237,7 @@ class MemoryMonitor {
             </head>
             <body>
                 <h1>Memory Monitor</h1>
-
+                
                 <div class="memory-card">
                     <h2>System Memory</h2>
                     <div class="memory-info">
@@ -259,7 +259,7 @@ class MemoryMonitor {
                         </div>
                     </div>
                 </div>
-
+                
                 <div class="memory-card">
                     <h2>Process Memory (VSCode Extension Host)</h2>
                     <div class="memory-info">
@@ -277,18 +277,18 @@ class MemoryMonitor {
                         </div>
                     </div>
                 </div>
-
+                
                 <div class="memory-card">
                     <h2>Memory Usage History</h2>
                     <div class="chart-container">
                         <canvas id="memoryChart"></canvas>
                     </div>
                 </div>
-
+                
                 <div class="memory-card">
                     ${processTable}
                 </div>
-
+                
                 <script>
                     // Create memory usage chart
                     const ctx = document.getElementById('memoryChart').getContext('2d');
@@ -323,10 +323,10 @@ class MemoryMonitor {
                             }
                         }
                     });
-
+                    
                     // Set up real-time updates
                     const vscode = acquireVsCodeApi();
-
+                    
                     // Request updates every 5 seconds
                     setInterval(() => {
                         vscode.postMessage({
@@ -340,4 +340,4 @@ class MemoryMonitor {
     }
 }
 
-module.exports = MemoryMonitor;
+module.exports = MemoryMonitor;

openhands/runtime/utils/vscode-extensions/memory-monitor/package.json

@@ -47,4 +47,4 @@
             ]
         }
     }
-}
+}