Actions AI Advisor
Version updated for https://github.com/ratibor78/actions-ai-advisor to version v1.
- This action is used across all versions by ? repositories.
Go to the GitHub Marketplace to find the latest changes.
Release notes
Actions AI Advisor v1
AI-powered GitHub Actions failure analysis - Stop digging through CI logs manually. Let AI explain failures for you.
π― First Stable Release
This is the first production-ready release of Actions AI Advisor!
β¨ Key Features
- π€ Intelligent Analysis - AI-powered root cause analysis directly in workflow summaries
- π Affected Files Detection - Automatically extracts and links to files mentioned in errors
- π Multi-Language Support - 10+ languages: Python, JS/TS, Go, Rust, Java, C#, PHP, Ruby, C/C++, Docker
- π Clickable File Links - Direct navigation to error locations with line numbers
- π° Cost Transparency - Shows token usage and estimated cost per analysis
- β‘ Fast Execution - Uses pre-built Docker images (5-10 seconds)
- π Provider Flexibility - OpenAI, Anthropic, OpenRouter, or self-hosted LLMs
π Quick Start
jobs:
test:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
- run: npm test
ai-advisor:
runs-on: ubuntu-latest
if: failure()
needs: test
permissions:
actions: read
steps:
- uses: ratibor78/actions-ai-advisor@v1
with:
github_token: ${{ github.token }}
api_key: ${{ secrets.OPENAI_API_KEY }}
provider: openai
model: gpt-4o-mini
π Documentation
π΅ Cost
~$0.0003-0.0008 per analysis (β1000-3000 analyses per $1)
π³ Docker Images
Available at: `ghcr.io/ratibor78/actions-ai-advisor`
Tags: `v1`
π What’s Changed
Features
- β¨ Complete project rename to actions-ai-advisor
- β‘ Optimized to use pre-built Docker images for fast execution
- π§ Snake_case input names for better compatibility
- π― Support for self-hosted LLM providers
- π¦ Multi-language file path extraction (10+ languages)
- π Clickable GitHub links to error locations
- π° Token usage and cost estimation
Technical Improvements
- π Python 3.12+ with modern tooling (uv, pydantic, httpx)
- π§ͺ 68 tests passing with comprehensive coverage
- π Full documentation and examples
- π Type-safe with mypy strict mode
- π¨ Linted with ruff