An autonomous agent designed to discover, analyze, and fix GitHub issues while maintaining strict security through a 5-layer guardrail system.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install github-issue-resolver
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 github-issue-resolver 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 GitHub Issue Resolver?
The GitHub Issue Resolver is a sophisticated autonomous agent that streamlines the entire lifecycle of software maintenance. By integrating directly with repository workflows, this tool helps developers manage technical debt by identifying high-impact issues and proposing verified fixes. It stands out in the ecosystem of Openclaw Skills by prioritizing safety and transparency through its robust internal security protocols.
The skill handles everything from repository cloning and branch management to automated testing and draft PR creation. It ensures that no dangerous operations, such as modifying protected branches or accessing secrets, occur without explicit oversight, making it a reliable partner for modern development teams. By leveraging standardized scripts for recommendation and auditing, it provides a consistent and professional experience for issue resolution.
GitHub Issue Resolver Use Cases
- Automatically discovering and scoring open bugs in a repository based on severity and impact.
- Developing code fixes for well-defined GitHub issues without manual context switching.
- Generating comprehensive test suites for new fixes to ensure regression-free deployments.
- Streamlining the pull request process by creating detailed draft PRs for human review.
- Managing multiple issues sequentially while ensuring an audit trail for every action taken.
How GitHub Issue Resolver Works
- Repository Discovery: The agent validates the target repository against security guardrails and uses a recommendation engine to score and present the best issues to tackle.
- Issue Analysis and Planning: Once an issue is selected, the agent locks the task, explores the codebase, and presents a detailed fix plan for user approval.
- Safe Implementation: After approval, the agent creates a dedicated branch and implements the fix within a secure sandbox environment.
- Validation and Testing: The agent runs existing test suites and writes new tests to verify the fix, with optional auto-rollback features if tests fail.
- PR Submission: Finally, the agent generates a change summary and creates a draft pull request on GitHub for final human verification.
GitHub Issue Resolver Setup
Ensure you have the GitHub CLI (gh) installed and authenticated. To begin using this component of Openclaw Skills, clone the repository and install dependencies:
# Initialize the environment
pip install -r requirements.txt
# Configure guardrails
cp guardrails.json.example guardrails.json
# Start the discovery process
python3 scripts/recommend.py <owner>/<repo>
GitHub Issue Resolver Data Schema & Taxonomy
The skill maintains a structured directory for operations and auditing to ensure full transparency.
| Path | Description |
|---|---|
/audit/ |
Contains JSON logs of every action and command executed for compliance. |
/references/ |
Documentation for guardrails and API scoring rubrics used by the agent. |
guardrails.json |
Configuration file defining allowed scopes, paths, and action gates. |
/tmp/openclaw-work/ |
Temporary sandbox directory for repository cloning and code manipulation. |
GitHub Issue Resolver Advanced Features
- 5-Layer Guardrail System: Comprehensive security checks covering scope validation, action gates, and command allowlists.
- Intelligent Scoring Engine: Automatically ranks issues by severity, effort, and freshness using the recommendation script.
- Automated Rollback: Configurable safety feature to revert changes instantly if tests do not pass during the validation phase.
- Audit Logging: Detailed tracking of all operations to ensure compliance and transparency within Openclaw Skills.
- Sandbox Execution: All shell commands are wrapped in a safe execution environment to prevent unauthorized system-level side effects.
SKILL.md
Loading
Related Openclaw Skills
Firecrawl
A comprehensive web search and scraping tool designed for AI agents to extract structured data from any website including JS-heavy pages.
PDF OCR using Gemini LLM
A sophisticated AI-powered tool that leverages Google Gemini to perform high-accuracy OCR on scanned or image-based PDF documents.
Pear iCloud Integration
A comprehensive suite of 27 MCP tools that provides read/write access to iCloud Calendar, Reminders, and Contacts through Pear's CalDAV and CardDAV bridge.
LLM Council Router
Route any prompt to the highest-performing LLM based on thousands of peer-reviewed council deliberations.
SeedDance AI Video Generation
A professional-grade integration for ByteDance's SeedDance API, enabling programmatic text-to-video and image-to-video generation.
Humanize AI Content
A specialized tool for transforming sterile AI-generated text into high-quality, natural-sounding, and publication-ready editorial content.
Featured*
