An automated Git commit analysis agent that verifies if code changes actually resolve specific bug reports or requirements.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install commit-reviewer
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install commit-reviewer using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Commit Reviewer?
Commit Reviewer is a specialized utility designed for the Openclaw Skills framework to bridge the gap between code commits and business requirements. It goes beyond simple diff viewing by performing a deep semantic analysis of one or multiple Git commit hashes against a provided bug description or feature requirement. By examining actual logic changes rather than just commit messages, it provides developers and tech leads with a rigorous assessment of whether a task is truly complete.
The tool is built to handle complex project structures, supporting automatic repository identification or explicit path definitions. It serves as an automated peer reviewer that highlights potential side effects, omissions, and risks, ensuring that high-quality code reaches production with fewer regressions.
Commit Reviewer Use Cases
- Verifying if a specific bug fix actually addresses the root cause described in a ticket.
- Cross-checking multiple commits to ensure all requirement points of a feature are covered.
- Performing quick risk assessments to identify potential side effects or logical omissions in a patch.
- Auditing frontend or UI changes to determine if manual verification is necessary due to visual complexity.
How Commit Reviewer Works
- The user triggers the skill by providing one or more Git commit hashes and an optional project path.
- The agent identifies the target repository using environment variables or directory scanning.
- If a requirement description is missing, the agent prompts the user to provide the context (bug report or feature spec).
- The script collects the full context of the commits, including the actual diff/patch data, using Git and internal bash utilities.
- The AI analyzes the code changes line-by-line against the requirements to form a technical conclusion.
- A structured report is generated, classifying each requirement point as Solved, Partially Solved, Unsolved, or Undeterminable.
Commit Reviewer Setup
Ensure you have the required binaries installed: bash, git, find, sed, grep, and sort. Configure your workspace root for better project identification.
# Set your workspace root for Openclaw Skills automatic scanning
export COMMIT_REVIEWER_WORK_ROOT=/your/workspace/root
# Optional: Set scan depth and patch line limits
export COMMIT_REVIEWER_SCAN_DEPTH=4
export COMMIT_REVIEWER_PATCH_LINES=1200
To use the skill, simply invoke the command within your AI agent interface:
/commit_reviewer <project_name> <commit_hash>
Commit Reviewer Data Schema & Taxonomy
Commit Reviewer organizes its analysis into a structured taxonomy to ensure clarity in Openclaw Skills reports:
| Section | Description |
|---|---|
| Commit List | List of all hashes analyzed in the session. |
| Requirement Breakdown | Atomic points extracted from the user's task description. |
| Point-by-Point Analysis | Detailed evidence, logic changes, and files modified for every requirement. |
| Overall Conclusion | A final status: Fixed, Partially Fixed, or Needs Further Verification. |
| Risk Metadata | Flags for side effects, style-only changes, or missing core logic. |
Commit Reviewer Advanced Features
- Multi-Commit Aggregation: Analyze a series of commits as a single logical unit of work.
- Contextual Project Discovery: Automatically locates the correct Git repository across deep directory structures.
- Intelligent Risk Flagging: Automatically detects when changes are purely cosmetic or if frontend logic requires manual QA.
- Custom Scan Depth: Configure how deep the Openclaw Skills engine should look for repositories using
COMMIT_REVIEWER_SCAN_DEPTH.
SKILL.md
Loading
Related Openclaw Skills
Equipment Maintenance Log
A specialized tool for tracking lab equipment calibration dates and generating automated maintenance reminders.
Queue State Diff
A specialized comparison tool for AI agents to analyze drift between JSON queue snapshots and detect operational regressions.
PydanticAI Tool System
A comprehensive system for registering, documenting, and managing function-calling tools within the PydanticAI framework.
Firecrawl CLI
A powerful command-line interface for scraping, crawling, and performing AI-powered structured data extraction from any website.
Web Profiler Bundle
A local-first profiling toolkit for tracking HTTP request timings, memory usage, and database query performance via the command line.
laolei-llm-midi
A professional AI music assistant that transforms natural language descriptions into multi-track, standard MIDI files for production and performance.
Featured*
