ultrathink-detective
⚡ PRIMARY TOOL for: 'comprehensive audit', 'deep analysis', 'full codebase review', 'multi-perspective investigation', 'complex questions'. Combines ALL detective perspectives (architect+developer+tester+debugger). Uses Opus model. REPLACES grep/glob entirely. Uses claudemem v0.3.0 AST with ALL commands (map, symbol, callers, callees, context). GREP/FIND/GLOB ARE FORBIDDEN.
npx skills add majiayu000/claude-skill-registry --skill ultrathink-detective --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
What it does
The skill orchestrates a multi-dimensional codebase audit using CLAUDEMEM AST commands to analyze architecture, exact locations, impact, dependencies, and full call chains, supplemented by semantic search and health checks. It emphasizes a holistic, evidence-based approach across architecture, implementation, tests, reliability, security, performance, and code health, with an emphasis on architecture-focused insights derived from high-PageRank symbols and full context analysis.
How it works
- It enforces AST structural analysis only, using claudemem v0.3.0+ tools (no grep/find/glob).
- It includes commands for architecture mapping (map), exact location pinpointing (symbol), impact analysis (callers), dependency tracing (callees), and full call chains (context).
- It uses semantic search (search) to discover broader patterns and integrates a multi-phase workflow covering architecture, implementation, tests, reliability, security, performance, and health.
- Setup steps validate claudemem v0.3.0, prompt for installation if missing, and check index freshness, with options to reindex.
When to use it
- Complex bugs spanning multiple systems
- Major refactoring decisions
- Technical debt assessment
- New developer onboarding
- Post-incident root cause analysis
- Architecture decision records
- Security audits
- Comprehensive code reviews
What it can touch
- Tools: claudemem commands (map, symbol, callers, callees, context, search) are invoked with --nologo and --raw to produce structured outputs.
- No external file touch is specified beyond the claudemem index management; all analysis prompts are CLI-based.
Caveats
- FORBIDDEN: GREP, FIND, GLOB usage per guidance.
- Requires claudemem v0.3.0+ and an index that may need refreshing; stale indexes may affect results.
- Analysis is AST-based and relies on the relationships captured by claudemem; results are only as accurate as the index and the codebase state.
# ⛔⛔⛔ CRITICAL: AST STRUCTURAL ANALYSIS ONLY ⛔⛔⛔ ``` ╔══════════════════════════════════════════════════════════════════════════════╗ ║ ║ ║ 🧠 THIS SKILL USES claudemem v0.3.0 AST ANALYSIS EXCLUSIVELY ║ ║ ║ ║ ❌ GREP IS FORBIDDEN ║ ║ ❌ FIND IS FORBIDDEN ║ ║ ❌ GLOB IS FORBIDDEN ║ ║ ║ ║ ✅ claudemem --nologo map "query" --raw FOR ARCHITECTURE ║ ║ ✅ claudemem --nologo symbol <name> --raw FOR EXACT LOCATIONS ║ ║ ✅ claudemem --nologo callers <name> --raw FOR IMPACT ANALYSIS ║ ║ ✅ claudemem --nologo callees <name> --raw FOR DEPENDENCY TRACING ║ ║ ✅ claudemem --nologo context <name> --raw FOR FULL CALL CHAIN ║ ║ ✅ claudemem --nologo search "query" --raw FOR SEMANTIC SEARCH ║ ║ ║ ║ ⭐ v0.3.0: ALL commands used for comprehensive multi-dimensional analysis ║ ║ ║ ╚══════════════════════════════════════════════════════════════════════════════╝ ``` # Ultrathink Detective Skill **Version:** 3.3.0 **Role:** Senior Principal Engineer / Tech Lead **Model:** Opus (for maximum reasoning depth) **Purpose:** Comprehensive multi-dimensional codebase investigation using ALL AST analysis commands with code health assessment ## Role Context You are investigating as a **Senior Principal Engineer**. Your analys
- Role Context
- Why Ultrathink Uses ALL Commands
- When to Use Ultrathink
- PHASE 0: MANDATORY SETUP (CANNOT BE SKIPPED)
- Step 1: Verify claudemem v0.3.0
- Step 2: If Not Installed → STOP
- Step 3: Check Index Status
- Step 3.5: Check Index Freshness
- Step 4: Index if Needed
- Multi-Dimensional Analysis Framework (v0.3.0)
- Dimension 1: Architecture (map command)
- Dimension 2: Implementation (callers/callees)
- Dimension 3: Test Coverage (callers analysis)
- Dimension 4: Reliability (context command)
which claudemem && claudemem --version Must be 0.3.0+ Check claudemem installation and index claudemem --version && ls -la .claudemem/index.db 2>/dev/null First check if index exists if [ ! -d ".claudemem" ] || [ ! -f ".claudemem/index.db" ]; then exit 1 fi Count files modified since last index if [ "$STALE_COUNT" -gt 0 ]; then
What does the ultrathink-detective skill do?
⚡ PRIMARY TOOL for: 'comprehensive audit', 'deep analysis', 'full codebase review', 'multi-perspective investigation', 'complex questions'. Combines ALL detective perspectives (architect+developer+tester+debugger). Uses Opus model. REPLACES grep/glob entirely. Uses claudemem v0.3.0 AST with ALL commands (map, symbol, callers, callees, context). GREP/FIND/GLOB ARE FORBIDDEN.
How do I install it?
Run `npx skills add majiayu000/claude-skill-registry --skill ultrathink-detective --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From majiayu000/claude-skill-registry, a repository with 534 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.