agentleFS
Sign inStart free
petrovicistefanpetrovicistefan/mcp-api-guardian/.cursor/skills/mcp-family-architecture/SKILL.md

mcp-family-architecture

Skill0 starsChanged 24 days agoMITView on GitHub

Explains MCP product-family architecture, control-plane vs local tools, security-guard cloud/Team vs control-plane quotas, and product IDs. Use when planning features, billing, hosted paths, or how products relate.

  • Reads credentials

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Not yet the live security-guard Team/feed backend

    line 38

≈ “Common” is an estimate: at least 1 in 5 of the 308,232 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

MCP Family Architecture

Layout & maturity

mcp-control-plane     → HTTP accounts/keys/quotas (MVP management API)
mcp-security-guard    → SHIPPED reference: plugin + CLI + GH Action + opt-in cloud/Team
mcp-dependency-doctor → local MVP
mcp-api-guardian      → local MVP
mcp-database-doctor   → local MVP
mcp-cost-optimizer    → local MVP
mcp-seo-auditor       → local MVP

Control-plane IDs (mcp-control-plane/src/store.js):
security-guard · dependency-doctor · api-guardian · database-doctor · cost-optimizer · seo-auditor

Two paid/hosted tracks (do not mix casually)

A. security-guard cloud (live, product-specific)

  • Endpoint default: https://mcp-security-cloud.petrovicistefan.workers.dev (override MCP_SECURITY_API_URL)
  • Key: MCP_SECURITY_API_KEY or plugin secure userConfig (empty default for Cowork)
  • Threat feed: POST /v1/check — package names/versions + SHA-256 tool/context hashes (+ plugin name/version). Never definitions/paths/secrets. Fail open.
  • Team: org policy sync, fleet inventory (names/versions only, admin gate + consent), approvals, keys/seats, dashboard, email/Slack alerts. Admin mutations are CLI only, not MCP tools.
  • Free local audit remains complete without any key.

B. control-plane (family MVP)

  • Accounts, hashed API keys (mcp_…), entitlements, monthly per-product quotas
  • Trusted hosted caller: GET /v1/entitlements, POST /v1/usage/consume
  • Idempotent requestId; 429 = no charge; 409 = unit conflict
  • Not yet the live security-guard Team/feed backend

Decision tree

  1. Change local free auditing/analysis? → Product repo; keep useful offline.
  2. security-guard Team / threat feed / privacy of payloads? → mcp-security-guard (cloud.ts, team.ts, PRIVACY.md).
  3. Cross-product account quota API? → mcp-control-plane.
  4. Unify billing later? → Explicit migration plan; do not silently point security-guard at control-plane consume.

Anti-patterns

  • Treating security-guard like an unpublished 0.1.0 zero-dep MVP
  • Locking free local features behind a key
  • Sending rich inventory (paths, env, tool text) to cloud
  • Exposing Team admin approve/revoke as MCP tools
  • Claiming unpaid README prices as live offers for MVP siblings

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free