
context-hunter
by MrLesk
Backlog.md - A tool for managing project collaboration between humans and AI Agents in a git ecosystem
Use Cases
Workflow Automation
Automate repetitive tasks and reduce manual work.
CI/CD Pipeline Setup
Build automated pipelines for build, test, and deploy.
Scheduled Execution
Automatically run tasks on a schedule.
SKILL.md
name: context-hunter description: Discover codebase patterns, conventions, and unwritten rules before making changes. Use when implementing features, fixing bugs, or refactoring code.
Context Hunter
Before writing code, investigate how similar problems are already solved in this codebase.
Before Implementation
Discover Existing Patterns
- Find analogous features: Search for code that solves similar problems. Study it before proposing your approach.
- Trace data flow: How does similar data move through the system? Note caching, validation, and error handling patterns.
- Identify utilities: Search for existing helpers before creating new ones.
Detect Unwritten Conventions
Look for implicit rules encoded in the codebase:
- Schema patterns:
deleted_atcolumns indicate soft-deletion. Audit columns indicate tracking requirements. - Naming patterns: Note consistency in
user_idvsuserIdvsuserID. - Test patterns: What's tested thoroughly reveals team priorities.
Verify Assumptions
- Run the test suite to understand current state
- Check linter and formatter configs
- Read recent commits in affected areas
- Examine database schemas for constraints
During Implementation
Match Existing Code
Your changes should be indistinguishable from existing code:
- Use the same patterns, abstractions, and utilities
- Follow the same error handling approach
- Respect module boundaries
- Match naming conventions exactly
Surface Concerns
When you discover conflicts between requirements and existing patterns:
- Ask clarifying questions before proceeding
- Flag risks you've identified
- Offer alternatives that align with codebase conventions
Checklist
Before proposing changes, confirm:
- Studied analogous features in the codebase
- Checked for reusable utilities
- Reviewed test patterns for similar functionality
- Noted naming and schema conventions
- Verified approach matches existing patterns
Score
Total Score
Based on repository quality metrics
SKILL.mdファイルが含まれている
ライセンスが設定されている
100文字以上の説明がある
GitHub Stars 1000以上
1ヶ月以内に更新
10回以上フォークされている
オープンIssueが50未満
プログラミング言語が設定されている
1つ以上のタグが設定されている
Reviews
Reviews coming soon
