Trace MCP
Routes code exploration to a code-graph MCP server instead of grep/read.
Test report
- Verdict
- Works with setup
- Score
- Tested
- Jul 13, 2026
- Environment
- Claude Code 2.x (agent harness)
- Upstream re-checked
- Jul 18, 2026 · ab3b075
Well-designed decision matrix (20+ tools mapped to tasks), but it's a pure routing skill for a separate trace-mcp MCP server — with that server not connected, Claude just falls back to grep anyway, which is exactly what happened in this test on a small repo.
Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works
- Installs cleanly 5/5
- Triggers reliably 5/5
- Output vs. baseline 6/10
- Docs & honesty 4/5
What Trace MCP does
A routing skill that redirects code navigation, impact analysis, and call-graph queries to the trace-mcp MCP server instead of native Read/Grep/Bash. Triggers whenever the agent needs to explore, understand, or modify a codebase that has trace-mcp indexed, and ships a token-efficiency decision matrix mapping common tasks (find a symbol, trace a request flow, check change impact) to specific tools.
How to install Trace MCP
git clone https://github.com/nikolai-vysotskyi/trace-mcp
cd trace-mcp
mkdir -p ~/.claude/skills
cp -r skills/trace-mcp ~/.claude/skills/trace-mcp
Skills live in ~/.claude/skills/ (global) or .claude/skills/
(per-project). Restart Claude Code after installing.
Commands — how to trigger Trace MCP
-
/trace-mcpRoutes code exploration to a code-graph MCP server instead of grep/read.
It also activates on plain-language prompts like these:
-
Find where this function is defined and everywhere it's called -
Trace the request flow through this codebase for the checkout endpoint -
What's the impact if I change this shared utility function
Frequently asked questions
- Is the Trace MCP skill free?
- Yes. The skill itself is free from nikolai-vysotskyi/trace-mcp. SkillProof publishes the install command and an independent test verdict at no cost.
- Does Trace MCP work with Claude Code?
- We tested it with Claude Code 2.x (agent harness) on Jul 13, 2026. Verdict: Works with setup. Well-designed decision matrix (20+ tools mapped to tasks), but it's a pure routing skill for a separate trace-mcp MCP server — with that server not connected, Claude just falls back to grep anyway, which is exactly what happened in this test on a small repo.
- What is the Trace MCP SkillProof Score?
- 8.0/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 6/10, docs & honesty 4/5.
- How do I install Trace MCP?
- Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
- Can I use Trace MCP with Cursor, Copilot, Gemini CLI, Codex or other AI tools?
- The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.