By lis186
Understand any codebase deeply with AI: generate architecture overviews, trace code flows and data paths, assess change impacts with migration plans, discover design patterns and examples, analyze dependencies for upgrades, and review git history for hotspots and risks using 6 agent skills and 28 commands.
> **Constitution**: [ANALYSIS_CONSTITUTION.md](../../../ANALYSIS_CONSTITUTION.md) v1.1
Complete YAML format specification for dependency upgrade analysis.
Advanced features, caching behavior, and best practices.
Complete verification checklist to ensure dependency analysis accuracy.
Complete step-by-step guide for dependency upgrade analysis.
Trace code execution paths and data flow. Use when user asks "how does X work", "what happens when X", "trace the flow of X", "where does data come from", or needs to understand feature implementation.
Quickly understand a new codebase's architecture, tech stack, and patterns. Use when user asks "what is this project", "project overview", "how is this codebase structured", "what tech stack", or when onboarding to a new codebase.
Analyze dependencies for upgrade planning and migration. Use when user asks "upgrade to X", "migrate from X to Y", "what breaks if we upgrade", "iOS 17 migration", "React 18 upgrade", or planning framework/SDK upgrades.
Analyze git history for hotspots, coupling, and knowledge distribution. Use when user asks "who knows this code", "what files change most", "hotspots", "bus factor", "knowledge silos", or needs to understand code evolution.
Analyze what code will be affected by changes. Use when user asks "what will break if I change X", "impact of changing X", "dependencies of X", "is it safe to modify X", or before making significant code changes.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
🌐 sourceatlas.io | English | 繁體中文
Get project overview in ~3 minutes by scanning <5% of files
A set of AI-powered slash commands for Claude Code that help you understand any codebase quickly.

SourceAtlas solves these in minutes, not days.
| Task | Before | After |
|---|---|---|
| Understand project architecture | 2-3 days | ~3-15 minutes ✓ |
| Find API implementation examples | Ask colleagues / random searching | ~seconds |
| Analyze impact of code changes | Manual tracking, hope for the best | ~1-2 minutes |
| Plan framework upgrade | Weeks of research | ~15-30 minutes |
| Find code hotspots & experts | Ask around | ~5-10 minutes |
✓ = benchmarked on 5 open-source projects. Others are estimates based on usage.
SourceAtlas uses information theory to prioritize high-entropy files (configs, READMEs, models) over implementation details. By scanning <5% of files, it achieves 70-80% understanding depth in minutes instead of days.

/sourceatlas:overview
In ~3 minutes, get: Tech stack, architecture patterns, project scale, code quality signals
Example output: Detects Swift 5.10 + MVVM + Coordinator, 12K files, test coverage level
/sourceatlas:pattern "api endpoint"
/sourceatlas:pattern "authentication"
/sourceatlas:pattern "database query"
In 0.1-30 seconds, get: 2-3 best example files with exact line numbers + implementation guide
Example output: Returns UserAPI.swift:45 with matching test file and implementation guide
221 patterns supported: MVVM, Networking, Core Data, React Hooks, Next.js API, Jetpack Compose, Vue Composable, FastAPI, Rails Controller...
/sourceatlas:impact "src/api/users.ts"
/sourceatlas:impact api "/api/users/{id}"
In 1-2 minutes, get: All dependents, Breaking Change risks, test coverage, migration steps
Example output: Lists 23 dependent files, identifies 5 breaking change risks
/sourceatlas:history
/sourceatlas:history src/
In 5-10 minutes, get: Hotspots (files that change constantly), Hidden Coupling, Knowledge Distribution
Example output: Shows PaymentService.swift with 47 changes, flags single-contributor bus factor risk
/sourceatlas:flow "user login"
/sourceatlas:flow "checkout process"
In 3-5 minutes, get: Entry points, complete execution path, boundary identification (API/DB/Auth/Payment)
Example output: Traces LoginViewController → AuthService → APIClient → TokenManager
/sourceatlas:deps "iOS 16 → 17"
/sourceatlas:deps "React 17 → 18"
/sourceatlas:deps "Python 3.11 → 3.12"
In 15-30 minutes, get: Deprecated APIs, version checks to remove, third-party compatibility, effort estimate
Example output: Migration checklist with removable version checks, deprecated APIs, effort estimate
Tested on 5 open-source projects: Firefox iOS, Discourse, Cal.com, Prefect, Thunderbird
| Command | Key Metric | Result | Report |
|---|---|---|---|
overview | Overall Accuracy | 93% (56/60) | ✓ |
pattern | Search Precision | 98.6% (7/7 cases) | ✓ |
impact | Internal Consistency | 100% (5/5 projects) | ✓ |
flow | Entry Point Detection | 100% (5/5 projects) | ✓ |
deps | Mode Detection | 100% (2/2 cases) | ✓ |
history | Hotspots Detection | 100% (Top 10) | ✓ |
npx claudepluginhub lis186/sourceatlas --plugin sourceatlasLocal codebase intelligence + change-safety gates for coding agents. Pre-indexes your repo's symbols, call graph, deps, and git history into SQLite (28 languages, 100% local, zero API keys), then exposes a lean 16-tool MCP core preset: graph-precise search, callers/impact blast radius, coupling, dead code, taint reachability, and pre-merge verify/critique gates that catch regressions, broken references, AI-slop duplication, and convention drift before they ship.
Smart codebase exploration with dependency mapping and structure analysis
Codebase exploration, refactoring, and quality analysis
Codebase intelligence for Claude Code. Indexes your codebase into five layers (Graph, Git, Docs, Decisions, Code Health) and exposes them through nine task-shaped MCP tools — so Claude understands architecture, ownership, hotspots, why code is built the way it is, and where the defect risk lives.
Graph-first code intelligence for AI agents. SurrealDB knowledge graph + 52 MCP tools replace Read/Grep/Glob with deterministic graph traversal. 80–95% fewer tokens on code context. Rust-native, fully local.
Semantic code intelligence -- symbol navigation, cross-reference analysis, and structural code understanding powered by tree-sitter