Skip to content
Open
Show file tree
Hide file tree
Changes from 16 commits
Commits
Show all changes
26 commits
Select commit Hold shift + click to select a range
fe9902d
feat(watch): Enhanced SSE monitoring with detailed event data
jack-arturo Jan 9, 2026
c30abcd
feat(stream): Add memory.associate event logging
jack-arturo Jan 9, 2026
e582888
fix(test): Update enrichment test for dict return type
jack-arturo Jan 9, 2026
03a3eac
feat(stream): Enhanced enrichment events with full context
jack-arturo Jan 9, 2026
b16112c
fix: Production bugs - datetime tz, OpenAI tokens, Qdrant race
jack-arturo Jan 9, 2026
ad425b5
fix: Correct token parameter for gpt-5 models (max_output_tokens)
jack-arturo Jan 9, 2026
3666301
Merge branch 'main' into feat/enhanced-sse-monitoring
jack-arturo Jan 10, 2026
00012a2
fix: Defensive coding improvements for automem_watch.py
jack-arturo Jan 10, 2026
a621599
fix: More defensive coding for automem_watch.py payloads
jack-arturo Jan 10, 2026
05fb0b4
fix: Final defensive coding fixes for automem_watch.py
jack-arturo Jan 10, 2026
9538ab8
refactor(watch): Use Event type alias for all event handler functions
jack-arturo Jan 10, 2026
ef85ca0
fix(watch): Defensive unpacking for semantic_neighbors
jack-arturo Jan 10, 2026
5726da8
fix(watch): SSE multi-line parsing, narrow exceptions, auth error han…
jack-arturo Jan 10, 2026
f51551e
fix: Log malformed event data in stream_events
jack-arturo Jan 10, 2026
59ea7c4
feat: add Codex auto-fix workflow for CI failures
jack-arturo Jan 10, 2026
63f668a
fix(watch): SSE multi-line parsing, narrow exceptions, auth error han…
jack-arturo Jan 10, 2026
824a774
feat: Add web-based SSE monitor dashboard
jack-arturo Jan 10, 2026
ab08c1b
feat(monitor): Add login screen with shared auth (Graph Viewer style)
jack-arturo Jan 10, 2026
e766140
fix(monitor): Use api_key param instead of token for SSE auth
jack-arturo Jan 10, 2026
78d002c
feat(monitor): Unified operations dashboard with status cards and inl…
jack-arturo Jan 10, 2026
6d5d8be
feat: Add CORS support for cross-origin clients
jack-arturo Jan 11, 2026
8eabae6
feat(stream): Add optional JSONL event logging with history hydration
jack-arturo Jan 11, 2026
37ad479
docs: Add event logging and streaming endpoints to CLAUDE.md
jack-arturo Jan 11, 2026
d1bb238
revert: Remove auto-fix workflow and CI prompt files
jack-arturo Jan 13, 2026
151305e
fix: address CodeRabbit review feedback
jack-arturo Jan 13, 2026
0615de7
fix(monitor): emit store/recall/associate events
jack-arturo Jan 13, 2026
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
36 changes: 36 additions & 0 deletions .github/codex/prompts/fix-ci.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,36 @@
# Fix CI Failures - AutoMem/AutoHub

You are fixing CI failures for the AutoMem/AutoHub project. This is a Python Flask application with:
- Flask + SSE (Server-Sent Events)
- FalkorDB (graph database)
- Qdrant (vector database)
- Docker/Docker Compose deployment
- pytest for testing

## Your Task

Analyze the test failures and CodeRabbit comments below, then make the **minimal changes** needed to fix them.

## Rules

1. **Be surgical** - Only change what's necessary to fix the specific issue
2. **Don't refactor** - Resist the urge to "improve" adjacent code
3. **Match existing style** - Follow the patterns already in the codebase
4. **Preserve API** - Don't change endpoints or response formats unless the fix requires it
5. **Keep tests passing** - Your fix should not break other tests

## Common Fixes

- **Import errors**: Check relative imports, ensure modules exist
- **Type hints**: Add proper typing for function signatures
- **Async issues**: Ensure proper async/await usage
- **Database errors**: Check connection handling and query syntax
- **API errors**: Validate request/response handling

## What NOT to Do

- Don't add new features
- Don't refactor working code
- Don't change Docker configuration unless required
- Don't modify environment variable handling unnecessarily
- Don't add excessive logging or comments
122 changes: 122 additions & 0 deletions .github/workflows/auto-fix.yml
Original file line number Diff line number Diff line change
@@ -0,0 +1,122 @@
name: Auto-Fix with Codex

on:
workflow_run:
workflows: ["CI"]
types: [completed]
workflow_dispatch:
inputs:
prompt:
description: 'Custom instructions (optional)'
required: false
type: string
branch:
description: 'Branch to fix (defaults to current)'
required: false
type: string

jobs:
auto-fix:
if: |
github.event_name == 'workflow_dispatch' ||
github.event.workflow_run.conclusion == 'failure'
runs-on: ubuntu-latest
permissions:
contents: write
pull-requests: write
actions: read

steps:
- name: Determine branch
id: branch
run: |
if [ "${{ github.event_name }}" == "workflow_dispatch" ]; then
BRANCH="${{ inputs.branch || github.ref_name }}"
else
BRANCH="${{ github.event.workflow_run.head_branch }}"
fi
echo "branch=$BRANCH" >> $GITHUB_OUTPUT
echo "🎯 Target branch: $BRANCH"

- uses: actions/checkout@v4
with:
ref: ${{ steps.branch.outputs.branch }}
fetch-depth: 0

- name: Setup Codex Auth
run: |
mkdir -p ~/.codex
echo '${{ secrets.CODEX_AUTH_JSON }}' > ~/.codex/auth.json

- name: Gather failure context
id: context
if: github.event_name == 'workflow_run'
run: |
echo "📋 Fetching CI failure logs..."
gh run view ${{ github.event.workflow_run.id }} --log-failed > /tmp/ci-failures.log 2>&1 || true

echo "📋 Fetching CodeRabbit comments..."
PR_NUMBER=$(gh pr list --head "${{ steps.branch.outputs.branch }}" --json number -q '.[0].number' 2>/dev/null || echo "")
if [ -n "$PR_NUMBER" ]; then
gh pr view "$PR_NUMBER" --json comments -q '.comments[] | select(.author.login == "coderabbitai") | .body' > /tmp/coderabbit.txt 2>&1 || true
fi

echo "# CI Failure Context" > /tmp/context.md
if [ -s /tmp/ci-failures.log ]; then
echo "## Test Failures" >> /tmp/context.md
echo '```' >> /tmp/context.md
tail -200 /tmp/ci-failures.log >> /tmp/context.md
echo '```' >> /tmp/context.md
fi
if [ -s /tmp/coderabbit.txt ]; then
echo "## CodeRabbit Comments" >> /tmp/context.md
cat /tmp/coderabbit.txt >> /tmp/context.md
fi
env:
GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}

- name: Prepare prompt
run: |
if [ -n "${{ inputs.prompt }}" ]; then
echo "${{ inputs.prompt }}" > /tmp/prompt.md
elif [ -f ".github/codex/prompts/fix-ci.md" ]; then
cat .github/codex/prompts/fix-ci.md > /tmp/prompt.md
else
cat > /tmp/prompt.md << 'PROMPT'
Fix the CI failures in this repository.
Make minimal, surgical changes only.
PROMPT
fi

if [ -f /tmp/context.md ]; then
echo "" >> /tmp/prompt.md
echo "---" >> /tmp/prompt.md
cat /tmp/context.md >> /tmp/prompt.md
fi

- name: Run Codex
uses: openai/codex-action@v1
with:
prompt-file: /tmp/prompt.md
codex-args: '["--full-auto"]'
sandbox: workspace-write
safety-strategy: drop-sudo

- name: Check for changes
id: changes
run: |
if git diff --quiet; then
echo "has_changes=false" >> $GITHUB_OUTPUT
else
echo "has_changes=true" >> $GITHUB_OUTPUT
git diff --stat
fi

- name: Commit and push
if: steps.changes.outputs.has_changes == 'true'
run: |
git config user.name "codex[bot]"
git config user.email "codex[bot]@users.noreply.github.com"
git add .
git commit -m "fix: auto-fix CI failures [skip ci]"
git push
Loading