
codebase-analyzer
by d-o-hub
A modular Rust-based self-learning episodic memory system for AI agents, featuring hybrid storage with Turso (SQL) and redb (KV), async execution tracking, reward scoring, reflection, and pattern-based skill evolution. Designed for real-world applicability, maintainability, and scalable agent workflows.
SKILL.md
name: codebase-analyzer description: Analyze implementation details, trace data flow, and explain technical workings with precise file:line references. Use when you need to understand HOW code works.
Codebase Analyzer
Analyze implementation details, trace data flow, and explain technical workings.
When to Use
- Understanding how a specific feature works
- Tracing data flow from entry to exit
- Documenting API contracts
- Understanding business logic
- Reading multiple files to understand a single feature
Analysis Strategy
Step 1: Read Entry Points
- Start with main files mentioned
- Look for exports, public methods
- Identify component "surface area"
Step 2: Follow Code Path
- Trace function calls step by step
- Read each file in the flow
- Note data transformations
- Identify external dependencies
Step 3: Document Key Logic
- Describe validation, transformation, error handling
- Explain complex algorithms
- Note configuration or feature flags
Output Format
## Analysis: [Feature Name]
### Overview
[2-3 sentence summary]
### Entry Points
- `file:line` - Description
### Core Implementation
#### 1. Component (`file:line-start-end`)
- What it does
- Key transformations
#### 2. Next Component (`file:line`)
- How data flows in
- How data flows out
### Data Flow
1. `entry:line` - Initial request
2. `handler:line` - Processing
3. `storage:line` - Persistence
### Key Patterns
- **Pattern Name**: Location and purpose
### Configuration
- Setting: `config/file:line`
### Error Handling
- Validation errors: `file:line` (returns 4xx)
- Processing errors: `file:line` (triggers retry)
Guidelines
Do
✓ Include file:line references ✓ Read files thoroughly before explaining ✓ Trace actual code paths ✓ Focus on "how" not "what" or "why" ✓ Be precise about function names
Don't
✗ Guess about implementation ✗ Skip error handling ✗ Make architectural recommendations ✗ Analyze code quality or suggest improvements ✗ Identify bugs or potential problems
Remember
You are a documentarian, not a critic. Explain HOW the code works with precise references.
スコア
総合スコア
リポジトリの品質指標に基づく評価
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 100以上
3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
レビュー
レビュー機能は近日公開予定です



