- Home
- Skills
- Testing & Debugging
- GitHub Issue-to-Pull-Request Agent
More screenshots
Works with the AI Agents you already use
GitHub Issue-to-Pull-Request Agent
Converts GitHub issues into repository-aware implementation plans, code changes, tests, and review-ready pull requests.
$7.99
GitHub Issue-to-Pull-Request Agent
Example session with this skill installed
Analyze issue #42 regarding duplicate webhook delivery and prepare a full implementation plan and code changes for the fix.
- Read your context and instructions
- Compiled the github issue-to-pull-request agent
Implementation Plan
- Move idempotency lookup before delivery creation in
webhook_service.py. - Add regression test in
tests/test_webhooks.py. - Update
CHANGELOG.md.
Code Changes
# snippet of fix...
PR Description
Closes #42. Fixes race condition in...
Connects securely to your tools. The creator never sees your data.
About this skill
The problem
GitHub issues are often implementation-ready. Developers waste time interpreting ambiguous requirements, rediscovering repository patterns, and writing repetitive PR descriptions instead of focusing on code.
What it does
- Extracts acceptance criteria and behavior models from GitHub issues and comments.
- Performs repository reconnaissance to identify impacted files and local coding conventions.
- Generates targeted code changes and automated tests that follow existing architecture.
- Performs multi-layered verification including security, privacy, and accessibility reviews.
- Produces complete pull request packages with titles, descriptions, and reviewer guidance.
Frameworks & tools
Works with GitHub, Cursor, Claude Code, and Copilot. Supports JavaScript, TypeScript, Python, Go, Rust, Java, Swift, Ruby, SQL, Terraform, and Docker.
Why this beats prompting it yourself
Standard prompts often result in scope creep, unrelated refactoring, or tests that validate implementation instead of behavior. This skill enforces a disciplined sequence from requirement modeling to evidence-based verification, ensuring minimal diffs and review-ready outputs.
Use cases
- Convert vague bug reports into reproducible test cases and verified fixes.
- Decompose large, complex issues into manageable, sequenced pull requests.
- Generate comprehensive API or UI feature implementations with migration notes.
- Audit existing diffs against original issue requirements for compliance.
Known limitations
Cannot independently guarantee deployment safety without repository access and executed CI evidence. High-impact architectural or security decisions require manual specialist approval.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Testing & Debugging.
Requires repository access for reconnaissance. Cannot bypass branch protections or finalize merges without human review. High-risk migrations require manual verification.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 3 months ago
Permissions
File Scopes
This skill uses file access to inspect GitHub issues, issue comments, source code, tests, package manifests, lock files, build files, configuration, schemas, migrations, API definitions, documentation, contribution guides, CODEOWNERS files, CI workflows, pull-request templates, commit conventions, architecture notes, product requirements, logs, stack traces, screenshots, performance reports, accessibility requirements, security requirements, support tickets, incident reports, and existing diffs. It uses write access to create or modify implementation files, test files, fixtures, mocks, configuration, documentation, migration files, changelog entries, implementation plans, verification records, commit summaries, reviewer notes, pull-request descriptions, risk assessments, rollback guidance, and follow-up issue specifications. Browser access is useful when the host environment permits authorized GitHub repository access, issue and pull-request review, current dependency-documentation review, framework-documentation validation, public API documentation research, security-advisory review, standards research, or compatibility validation. The skill should clearly distinguish: Verified Requirement: Explicitly supported by the issue, an authorized maintainer decision, an approved specification, or current repository behavior. Repository Evidence: Confirmed through inspected source code, tests, configuration, schema, documentation, or repository conventions. Assumption: A provisional interpretation that is not yet confirmed. Recommended: A proposed implementation, test, migration, rollout, or review approach based on available evidence. Open Question: A missing decision that affects correctness, scope, security, privacy, compatibility, architecture, accessibility, performance, migration, billing, or user behavior. Executed Verification: A command or check that was actually run and has recorded evidence. Not Run: A validation step that was identified but not executed. Blocked: A step that could not be completed because of missing access, dependencies, environment, evidence, or approval. The agent must never invent repository files, API contracts, database constraints, environment variables, test results, command output, CI results, deployment outcomes, or pull-request readiness. The agent must never expose or hard-code passwords, tokens, private keys, credentials, customer data, private production records, or internal secrets. The agent should preserve existing architecture and repository conventions unless the issue explicitly requires a justified change. The agent should implement only the verified issue requirements and the minimum supporting changes required for correctness, testability, maintainability, compatibility, and documentation. Generated files should not be edited directly when repository instructions require modifying the source definition and regenerating the output. Baseline test failures must be distinguished from failures introduced by the proposed change. A pull request should not be described as merge-ready when required validation, review, migration analysis, or security approval is missing. Use fictional, anonymized, masked, or approved data in public examples, tests, fixtures, documentation, and screenshots.