This analyzes your active MCP servers and calculates how much token overhead each one adds to every single API request. The insight here is simple but easy to miss: a server with 20 tools burns 2-4K tokens per request whether you use those tools or not. Run this when sessions feel sluggish or before adding new servers. It counts tools per server, checks which ones you actually use, identifies overlaps, and recommends what to disable. The output includes estimated savings in both tokens and dollars. Good for anyone running more than a handful of MCP servers who wants to understand why their context window fills up faster than expected.

skills.sh installs
169
GitHub stars
2.9K
last updated
Jun 8, 2026

Install

$ npx -y skills add rohitg00/pro-workflow --skill mcp-audit --agent claude-code

Installs into .claude/skills of the current project.

Files

MCP Audit

Analyze MCP server overhead and recommend cleanup.

Trigger

Use when:

  • Sessions feel slow or expensive
  • Adding a new MCP server
  • Context fills up quickly
  • Reviewing project configuration

Key Insight

Each MCP server adds ALL its tool descriptions to every API request. A server with 20 tools adds ~2K-4K tokens per request, regardless of whether you use those tools.

Audit Steps

Step 1: List Active Servers

Check all MCP configurations:

cat .claude/settings.json 2>/dev/null | grep -A 50 "mcpServers"
cat ~/.claude/settings.json 2>/dev/null | grep -A 50 "mcpServers"

Step 2: Count Tools Per Server

For each server, estimate token overhead:

  • 1-5 tools: ~200-500 tokens (low overhead)
  • 6-15 tools: ~500-1500 tokens (moderate)
  • 16-30 tools: ~1500-3000 tokens (high)
  • 30+ tools: ~3000+ tokens (excessive — consider tool filtering)

Step 3: Check Usage

Questions to ask:

  • Which servers were actually used this session?
  • Which servers haven't been used in 7+ days?
  • Are there servers with overlapping functionality?
  • Are there servers only needed for specific tasks?

Step 4: Recommend Actions

Sponsors

Details

Repository
github.com/rohitg00/pro-workflow
Categories
Security, AI & Agent Building, Automation & Workflows
First seen
Jun 11, 2026

More from rohitg00/pro-workflow

All 35

Recommended

More Security
1daffy0208 avatarcodex-review-workflowdaffy0208/ai-dev-standardsAutomated code review workflow using OpenAI Codex CLI. Implements iterative fix-and-review cycles until code passes validation or reaches iteration limit. Use w166skills.sh installs
2cinience avataralicloud-database-analyticdb-mysql-testcinience/alicloud-skillsSmoke test for alicloud-database-analyticdb-mysql. Validate minimal authentication, API reachability, and one read-only query path.164skills.sh installs
3jwilger avatarcode-reviewjwilger/agent-skillsThree-stage code review protocol covering spec compliance, code quality, and domain integrity. Use this skill whenever the user asks to review code, prepare or 163skills.sh installs
4jezweb avatardependency-auditjezweb/claude-skillsComprehensive dependency health auditing for JavaScript/TypeScript projects. Run npm audit, detect outdated packages, check for security advisories, and verify 160skills.sh installs
5cinience avataralicloud-media-live-testcinience/alicloud-skillsSmoke test for alicloud-media-live. Validate minimal authentication, API reachability, and one read-only query path.159skills.sh installs
6cinience avataralicloud-ai-recommend-airec-testcinience/alicloud-skillsSmoke test for alicloud-ai-recommend-airec. Validate minimal authentication, API reachability, and one read-only query path.158skills.sh installs