## Task
The SmartDecisionMakerBlock is a specialized block in a graph-based
system that leverages a language model (LLM) to make intelligent
decisions about which tools or functions to invoke based on a
user-provided prompt. It is designed to process input data, interact
with a language model, and dynamically determine the appropriate tools
to call from a set of available options, making it a powerful component
for AI-driven workflows.
## How It Works in Practice
- **Scenario:** Imagine a workflow where a user inputs, "Send an email
to John about the meeting." The SmartDecisionMakerBlock is connected to
tools like send_email, schedule_meeting, and search_contacts.
- **Execution:**
1. The block receives the prompt and system instructions (e.g., "Choose
a function to call").
2.It identifies the available tools from the graph and constructs their
signatures (e.g., send_email(recipient, subject, body)).
3. The LLM analyzes the prompt and decides to call send_email with
arguments like recipient: "John", subject: "Meeting", body: "Let’s
discuss...".
4. The block yields these tool-specific outputs, which can be picked up
by downstream nodes to execute the email-sending action.
## Changes 🏗️
- Add the Smart Decision Maker (SDM) block.
- Break circular imports in integration code.
## Work in Progress
⚠️ **Important note this is a temporary UX for the system - UX will be
addressed in a future PR** ⚠️
### Current Status
I’m currently focused on the smart decision logic. The main additions in
the ongoing PR include:
- Defining function signatures for OpenAI function-calling schemas based
on node links and the linked blocks.
- Adding tests for function signature generation.
- Force all tool calls to be made via an agent. (Need to uncomment)
- Restrict each tool call entry to a single node.
- simplify the output emission process, to emit each parameter one at a
time.
- Change test to use agents and hardcode output how I think it should
work to test it does actually work
- Hook up openai, in a simplified way, to test the function calling
(mock for testing)
- Once all the above is working, use credentials system and build of
llm.py
### What’s Next
- Review Process
### Reviewers Phase 1
This PR is now ready for review, during the first phase of reviews I'm
looking for comments on approach and logic.
Out of scope: code style and organization at this stage
### Reviewers Phase 2
Once we are all happy with the approach and logic. We can open the
review process to general code quality and nits, to be considered.
---------
Co-authored-by: Zamil Majdy <zamil.majdy@agpt.co>
- Resolves#8780
- Part of #8774
### Changes 🏗️
- Add new UI components
- Add `/agents/[id]` page, with sub-components:
- `AgentRunsSelectorList`
- `AgentRunSummaryCard`
- `AgentRunStatusChip`
- `AgentRunDetailsView`
- `AgentRunDraftView`
- `AgentScheduleDetailsView`
Backend improvements:
- Improve output of execution-related API endpoints: return
`GraphExecution` instead of `NodeExecutionResult[]`
- Reduce log spam from Prisma in tests
General frontend improvements:
- Hide nav link names on smaller screens to prevent navbar overflow
- Clean up styling and fix sizing of `agptui/Button`
Technical frontend improvements:
- Fix tailwind config size increments
- Rename `font-poppin` -> `font-poppins`
- Clean up component implementations and usages
- Yeet all occurrences of `variant="default"`
- Remove `default` button variant as duplicate of `outline`; make
`outline` the default
- Fix minor typing issues
DX:
- Add front end type-check step to `pre-commit` config
- Fix logging setup in conftest.py
### Checklist 📋
#### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
- `/agents/[id]` (new)
- Go to page -> list of runs loads
- Create new run -> runs; all I/O is visible
- Click "Run again" -> runs again with same input
- `/monitoring` (existing)
- Go to page -> everything loads
- Selecting agents and agent runs works
---------
Co-authored-by: Nicholas Tindle <nicktindle@outlook.com>
Co-authored-by: Nicholas Tindle <nicholas.tindle@agpt.co>
Co-authored-by: Swifty <craigswift13@gmail.com>
Co-authored-by: Zamil Majdy <zamil.majdy@agpt.co>
- Blocked by #9267
This re-introduces changes from the following PRs with fixes:
- #9218
- #9211
### Changes 🏗️
- See #9218
- See #9211
Fixes:
- Fix Prisma query statements in `v2.library.db`
- Fix creation of (library) agents
- Fix test cleanup of (library) agents
- Fix handling and passing of `node_input` parameters
### Checklist 📋
#### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
- [x] Create & run a new agent
- [x] Update & run an existing agent
Currently it's only possible to open latest graph from monitor and see
the node execution results only when manually running. This PR adds
ability to open running and finished graphs in builder.
### Changes 🏗️
Builder now handles graph version and execution ID in addition to graph
ID when opening a graph. When an execution ID is provided, node
execution results are fetched and subscribed to in real time. This makes
it possible to open a graph that is already executing and see both
existing node execution data and real-time updates (if it's still
running).
- Use graph version and execution id on the builder page and in
`useAgentGraph`
- Use graph version on the `execute_graph` endpoint
- Use graph version on the websockets to distinguish between versions
- Move `formatEdgeID` to utils; it's used in `useAgentGraph.ts` and in
`Flow.tsx`
### Checklist 📋
#### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
- [x] Opening finished execution restores node results
- [x] Opening running execution restores results and continues to run
properly
- [x] Results are separate for each graph across multiple tabs
#### For configuration changes:
- [ ] `.env.example` is updated or already compatible with my changes
- [ ] `docker-compose.yml` is updated or already compatible with my
changes
- [ ] I have included a list of my configuration changes in the PR
description (under **Changes**)
<details>
<summary>Examples of configuration changes</summary>
- Changing ports
- Adding new services that need to communicate with each other
- Secrets or environment variable changes
- New or infrastructure changes such as databases
</details>
---------
Co-authored-by: Zamil Majdy <zamil.majdy@agpt.co>
We want to have some more intelligent and less user managed memory
methods, so we add mem0
### Changes 🏗️
- Adds user_id to kwargs for blocks
- Add mem0 blocks
<!-- Concisely describe all of the changes made in this pull request:
-->
### Checklist 📋
- [x] document adding user_id to kwargs for blocks
- [x] Add run and agent Id as optional checkboxes that will be passed
down to mem0
#### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [ ] I have tested my changes according to the test plan:
<!-- Put your test plan here: -->
- [ ] Build and submit an agent to @Torantulino and the marketplace for
a personal AI tutor based on recommendations from the mem0 team
---------
Co-authored-by: Aarushi <50577581+aarushik93@users.noreply.github.com>
Co-authored-by: Zamil Majdy <zamil.majdy@agpt.co>
This re-introduces PR #9179 with some fixes.
This PR enables the execution of store agents even if they are not owned
by the user. Key changes include handling store-listed agents in the
`get_graph` logic, improving execution flow, and ensuring
version-specific handling. These updates support more flexible agent
execution.
### Changes 🏗️
(copied from #9179)
- **Graph Retrieval:** Updated `get_graph` to check store listings for
agents not owned by the user.
- **Version Handling:** Added `graph_version` to execution methods for
consistent version-specific execution.
- **Execution Flow:** Refactored `scheduler.py`, `rest_api.py`, and
other modules for clearer logic and better maintainability.
- **Testing:** Updated `test_manager.py` and other test cases to
validate execution of store-listed agents added test for accessing graph
Out-of-scope changes:
- Add logic to pretty-print Pydantic validation error responses to
backend API client in frontend
---------
Co-authored-by: Nicholas Tindle <nicktindle@outlook.com>
Co-authored-by: Nicholas Tindle <nicholas.tindle@agpt.co>
Co-authored-by: Swifty <craigswift13@gmail.com>
Co-authored-by: Zamil Majdy <zamil.majdy@agpt.co>
We want to allow external api calls against our platform
We also want to keep it sep from internal platform calls for dev ex,
security and scale seperation of concerns
### Changes 🏗️
This PR adds the required external routes
It mounts the new routes on the same app
Infra PR will seprate routing and domains
### Checklist 📋
#### For code changes:
- [ ] I have clearly listed my changes in the PR description
- [ ] I have made a test plan
- [ ] I have tested my changes according to the test plan:
<!-- Put your test plan here: -->
- [ ] ...
<details>
<summary>Example test plan</summary>
- [ ] Create from scratch and execute an agent with at least 3 blocks
- [ ] Import an agent from file upload, and confirm it executes
correctly
- [ ] Upload agent to marketplace
- [ ] Import an agent from marketplace and confirm it executes correctly
- [ ] Edit an agent from monitor, and confirm it executes correctly
</details>
#### For configuration changes:
- [ ] `.env.example` is updated or already compatible with my changes
- [ ] `docker-compose.yml` is updated or already compatible with my
changes
- [ ] I have included a list of my configuration changes in the PR
description (under **Changes**)
<details>
<summary>Examples of configuration changes</summary>
- Changing ports
- Adding new services that need to communicate with each other
- Secrets or environment variable changes
- New or infrastructure changes such as databases
</details>
- **Revert "feature(platform): Implement library add, update, remove,
archive functionality (#9218)"**
- **Revert "feat(backend): Add Support for Managing Agent Presets with
Pagination and Soft Delete (#9211)"**
These PRs contain untested changes to DB functions and cause issues in
production.
### Description
This PR enables the execution of store agents even if they are not owned
by the user. Key changes include handling store-listed agents in the
`get_graph` logic, improving execution flow, and ensuring
version-specific handling. These updates support more flexible agent
execution.
### Changes 🏗️
- **Graph Retrieval:** Updated `get_graph` to check store listings for
agents not owned by the user.
- **Version Handling:** Added `graph_version` to execution methods for
consistent version-specific execution.
- **Execution Flow:** Refactored `scheduler.py`, `rest_api.py`, and
other modules for clearer logic and better maintainability.
- **Testing:** Updated `test_manager.py` and other test cases to
validate execution of store-listed agents added test for accessing graph
---------
Co-authored-by: Reinier van der Leer <pwuts@agpt.co>
Co-authored-by: Zamil Majdy <zamil.majdy@agpt.co>
Python format uses `{Variable}` as the variable placeholder, while Jinja
uses `{{Variable}}` as its default.
Jinja is used as the main templating engine on the system, but the
Python format version is still maintained for backward compatibility.
However, the backward compatibility support can cause a side effect
while passing JSON string value into the block that uses it:
https://github.com/Significant-Gravitas/AutoGPT/issues/9194
### Changes 🏗️
* Use `{{Variable}}` place holder format and removed `{Variable}`
support in these blocks:
- '363ae599-353e-4804-937e-b2ee3cef3da4', -- AgentOutputBlock
- 'db7d8f02-2f44-4c55-ab7a-eae0941f0c30', -- FillTextTemplateBlock
- '1f292d4a-41a4-4977-9684-7c8d560b9f91', -- AITextGeneratorBlock
- 'ed55ac19-356e-4243-a6cb-bc599e9b716f' --
AIStructuredResponseGeneratorBlock
* Add Jinja templating support on `AITextGeneratorBlock` &
`AIStructuredResponseGeneratorBlock`
* Migrated the existing database content to prevent breaking changes.
### Checklist 📋
#### For code changes:
- [ ] I have clearly listed my changes in the PR description
- [ ] I have made a test plan
- [ ] I have tested my changes according to the test plan:
<!-- Put your test plan here: -->
- [ ] ...
<details>
<summary>Example test plan</summary>
- [ ] Create from scratch and execute an agent with at least 3 blocks
- [ ] Import an agent from file upload, and confirm it executes
correctly
- [ ] Upload agent to marketplace
- [ ] Import an agent from marketplace and confirm it executes correctly
- [ ] Edit an agent from monitor, and confirm it executes correctly
</details>
#### For configuration changes:
- [ ] `.env.example` is updated or already compatible with my changes
- [ ] `docker-compose.yml` is updated or already compatible with my
changes
- [ ] I have included a list of my configuration changes in the PR
description (under **Changes**)
<details>
<summary>Examples of configuration changes</summary>
- Changing ports
- Adding new services that need to communicate with each other
- Secrets or environment variable changes
- New or infrastructure changes such as databases
</details>
This is a follow-up of
https://github.com/Significant-Gravitas/AutoGPT/pull/8752
There are several APIs and functions related to graph execution that are
unused now.
There is also confusion about the name of `GraphExecution` that exists
in graph.py & execution.py.
### Changes 🏗️
* Renamed `GraphExecution` in `execution.py` to `GraphExecutionEntry`,
this is only used as a queue entry for execution.
* Removed unused `get_graph_execution` & `list_executions` in
`execution.py`.
* Removed `with_run` option on `get_graph` function in `graph.py`.
* Removed `GraphMetaWithRuns`
* Removed exposed functions only for testing.
* Removed `executions` fields in Graph model.
### Checklist 📋
#### For code changes:
- [ ] I have clearly listed my changes in the PR description
- [ ] I have made a test plan
- [ ] I have tested my changes according to the test plan:
<!-- Put your test plan here: -->
- [ ] ...
<details>
<summary>Example test plan</summary>
- [ ] Create from scratch and execute an agent with at least 3 blocks
- [ ] Import an agent from file upload, and confirm it executes
correctly
- [ ] Upload agent to marketplace
- [ ] Import an agent from marketplace and confirm it executes correctly
- [ ] Edit an agent from monitor, and confirm it executes correctly
</details>
#### For configuration changes:
- [ ] `.env.example` is updated or already compatible with my changes
- [ ] `docker-compose.yml` is updated or already compatible with my
changes
- [ ] I have included a list of my configuration changes in the PR
description (under **Changes**)
<details>
<summary>Examples of configuration changes</summary>
- Changing ports
- Adding new services that need to communicate with each other
- Secrets or environment variable changes
- New or infrastructure changes such as databases
</details>
---------
Co-authored-by: Krzysztof Czerwinski <34861343+kcze@users.noreply.github.com>
* fix(backend): Add execution persistence for execution scheduler service
* scheduler REST API cleanup
* Fix to binary
* Adapt UI with new API
* Remove schedule.py
* Remove unused class
* Fix linting
- feat(backend/executor): Change credential injection mechanism to acquire credentials from `AgentServer` just before execution
- Also locks the credentials for the duration of the execution
- feat(backend/server): Add thread-safe `IntegrationCredentialsManager` to handle and synchronize credentials-related operations
- feat(libs): Add mutexes to `SupabaseIntegrationCredentialsStore` to ensure thread-safety
Also:
- feat(backend): Added Pydantic model (de)serialization support to `@expose` decorator
Refactorings:
- refactor(backend, libs): Move `KeyedMutex` to `autogpt_libs.utils.synchronize`
- refactor(backend/server): Make `backend.server.integrations` module with `router`, `creds_manager`, and `utils` in it
Restructuring the Repo to make it clear the difference between classic autogpt and the autogpt platform:
* Move the "classic" projects `autogpt`, `forge`, `frontend`, and `benchmark` into a `classic` folder
* Also rename `autogpt` to `original_autogpt` for absolute clarity
* Rename `rnd/` to `autogpt_platform/`
* `rnd/autogpt_builder` -> `autogpt_platform/frontend`
* `rnd/autogpt_server` -> `autogpt_platform/backend`
* Adjust any paths accordingly
