* initial commit
* removes assesment from docs
* resolves automated review comments
* resolves lint , type , tests , refactors , and submits
* solves : why do we have to lint the tests xD
* adds greptile fixes
* solves a type error
* solves a ci error
* refactors auths
* solves a failing test after i pulled from main lol
* solves a failing test after i pulled from main lol
* resolves token naming issue to comply with better practices when using hf / huggingface
* fixes curly lints !
* fixes failing tests for google api from main
* solve merge conflicts
* solve failing tests with a defensive check 'undefined' openrouterapi key
* fix: preserve Hugging Face auth-choice intent and token behavior (#13472) (thanks @Josephrp)
* test: resolve auth-choice cherry-pick conflict cleanup (#13472)
---------
Co-authored-by: Cursor <cursoragent@cursor.com>
Co-authored-by: Peter Steinberger <steipete@gmail.com>
* Agents: allow gpt-5.3-codex-spark in fallback and thinking
* Fix: model picker issue for openai-codex/gpt-5.3-codex-spark
Fixed an issue in the model picker.
Skills install runs package manager install commands (npm, pnpm, yarn,
bun) without --ignore-scripts, allowing malicious npm packages to
execute arbitrary code via postinstall/preinstall lifecycle scripts
during global installation.
This is inconsistent with the security fix in commit 92702af7a which
added --ignore-scripts to both plugin installs (src/plugins/install.ts)
and hook installs (src/hooks/install.ts). Skills install was overlooked
in that change.
Global install (-g) is particularly dangerous as scripts execute with
the user's full permissions and can modify globally-accessible binaries.
The UsageAccumulator sums cacheRead/cacheWrite across all API calls
within a single turn. With Anthropic prompt caching, each call reports
cacheRead ≈ current_context_size, so after N tool-call round-trips the
accumulated total becomes N × actual_context, which gets clamped to
contextWindow (200k) by deriveSessionTotalTokens().
Fix: track the most recent API call's cache fields separately and use
them in toNormalizedUsage() for context-size reporting. This makes
/status Context display accurate while preserving accumulated output
token counts.
Fixes#13698Fixes#13782
Co-authored-by: akari-musubi <259925157+akari-musubi@users.noreply.github.com>
* fix: prevent FD leaks in child process cleanup
- Destroy stdio streams (stdin/stdout/stderr) after process exit
- Remove event listeners to prevent memory leaks
- Clean up child process reference in moveToFinished()
- Also fixes model override handling in agent.ts
Fixes EBADF errors caused by accumulating file descriptors
from sub-agent spawns.
* Fix: allow stdin destroy in process registry cleanup
---------
Co-authored-by: Tak Hoffman <781889+Takhoffman@users.noreply.github.com>
Two fixes for Google Antigravity (Cloud Code Assist) reliability:
1. Forward-compat model fallback: pi-ai's model registry doesn't include
claude-opus-4-6-thinking. Add resolveAntigravityOpus46ForwardCompatModel()
that clones the opus-4-5 template so the correct api ("google-gemini-cli")
and baseUrl are preserved. Fixes#13765.
2. Fix thinking.signature rejection: The API returns Claude thinking blocks
without signatures, then rejects them on replay. The existing sanitizer
strips unsigned blocks, but the orphaned-user-message path in attempt.ts
bypassed it by reading directly from disk. Now applies
sanitizeAntigravityThinkingBlocks at that code path.
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Move appendCacheTtlTimestamp() to after prompt + compaction retry
completes instead of before. The previous placement inserted a custom
entry (openclaw.cache-ttl) between compaction and the next prompt,
which broke pi-coding-agent's prepareCompaction() guard — the guard
only checks if the last entry is type 'compaction', and the cache-ttl
custom entry made it type 'custom', allowing an immediate second
compaction at very low token counts (e.g. 5,545 tokens) that nuked
all preserved context.
Fixes#9282
Relates to #12170
