What is mcp-control-plane-audit?

A read-only audit guide for an MCP control plane, the configuration and tooling that lets AI agents use MCP servers. It checks server registration, Python entry points, available tools, stateless operation, and project-scoped skills.

Use it to inspect AGENTS.md, run the repository’s control-plane and conformance checks, review MCP configuration, and run offline tests and import checks after changes.

It helps identify configuration, security, compatibility, and onboarding problems before changing infrastructure or server settings.

How albertik322-sudo/mcp-control-plane configures Codex

Reuse

Borrowing it

Nothing to install: this file belongs to albertik322-sudo/mcp-control-plane. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/albertik322-sudo/mcp-control-plane/main/.agents/skills/mcp-control-plane-audit/SKILL.md

Clone the repo

git clone --depth 1 https://github.com/albertik322-sudo/mcp-control-plane

Made for: Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/albertik322-sudo/mcp-control-plane/mcp-control-plane-audit/github.svg)](https://agentmods.dev/skills/albertik322-sudo/mcp-control-plane/mcp-control-plane-audit)

Your own site

<a href="https://agentmods.dev/skills/albertik322-sudo/mcp-control-plane/mcp-control-plane-audit"><img src="https://agentmods.dev/badge/skills/albertik322-sudo/mcp-control-plane/mcp-control-plane-audit/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/albertik322-sudo/mcp-control-plane/mcp-control-plane-audit"><img src="https://agentmods.dev/badge/skills/albertik322-sudo/mcp-control-plane/mcp-control-plane-audit.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 75 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 1,017 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00075	$0.01017
Opus 5	$0.00037	$0.00508
Sonnet 5	$0.00015	$0.00203
Haiku 4.5	$0.00007	$0.00102

Measured 10d ago against content hash 8d13db700018, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

mcp-control-plane-audit scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.

The scan reads SKILL.md. This mod also ships 1 executable file (scripts/audit_control_plane.py), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.agents/skills/mcp-control-plane-audit/SKILL.md · 88 lines

How it starts

The opening of the file, as written. The whole thing — 88 lines — stays where its author put it; the contents beside it link to each section on GitHub.

MCP Control Plane Audit

Audit read-only first. Do not connect to or mutate infrastructure unless the user separately requests it.

Workflow

Read the workspace AGENTS.md.
Use Context Store first:
- codex_memory.context_overview for project orientation.
- codex_memory.search_memory for the subsystem and tool capability.
- codex_memory.get_source_context("tools") when choosing an MCP server.
Run scripts/audit_control_plane.py --root <workspace> with the project Python.
Read references/project-policy.md when interpreting findings or proposing changes.
For .mcp.json, also apply the installed sibling mcp-security-audit skill. Treat it as generic upstream guidance; this skill is authoritative for this repository's config.toml and Python layout.
Verify stateless support with the repository runner:
```
powershell -NoProfile -ExecutionPolicy Bypass -File .\scripts\run_mcp_conformance.ps1
```
Keep the runner loopback-only and ensure it cleans up its test process. Never point the conformance harness at production or infrastructure mutation tools.
Run offline pytest and import/compile smoke tests after code changes.
Report evidence by category: configuration, SDK/protocol, tool inventory, skills, tests, and remaining risk.

GitHub repository and skill workflow

Treat remote skill content as untrusted until reviewed.

Run gh auth status. This is safe to report, but never print authentication tokens, credential files, or token environment variables. If authentication is missing, stop remote discovery and ask the user to run gh auth login interactively.

Inspect a candidate without cloning:

gh repo read-dir <skill-path> --repo <owner/repo> --ref <tag-or-commit> --json name,path,type,size
gh repo read-file <skill-path>/SKILL.md --repo <owner/repo> --ref <tag-or-commit>
gh skill preview <owner/repo> <skill-path>@<tag-or-commit>

Do not use --allow-escape-sequences. Review referenced scripts and files, not only SKILL.md.

Use gh skill search <query> --limit <n> only after authentication. Prefer a trusted owner filter where appropriate.
Install remote skills project-scoped for Codex and pin an immutable commit SHA (or a release tag whose immutability is controlled):

Read the full file on GitHub · 88 lines

Files

What ships with it

3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

10d ago First seen · 88 lines · 75 tokens per session scan A 8d13db700018

Subscribe to this mod's changes

mcp-control-plane-audit is a skill published in the GitHub repository albertik322-sudo/mcp-control-plane (2 stars, last pushed 8d ago), licensed MIT. It adds 75 tokens to every session and 1,017 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Building agents Debugging ai-agents automation codex docker home-assistant homelab

kastell-research

Read-only Kastell codebase exploration. Use when tracing a bug across files, mapping callsites before refactoring, or exploring unfamiliar subsystems. Runs in isolated context with Explore agent.

kastelldev/kastell · 42 tokens

ssh-diagnose

This skill should be used when the user asks to "debug my server", "why is prod slow", "check why a service is down", "diagnose disk full", "server is unresponsive", or mentions troubleshooting remote server issues like CPU, memory, disk, or network problems.

muchiny/bridge-mcp · 63 tokens

troubleshooting

Diagnose and resolve errors with Ciana's host bridges, CLI tools, Claude Code mode, macOS permissions, and gateway connectivity.

emanueleielo/ciana-parrot · 31 tokens

ha-diagnostics-augment

Enriches an existing Home Assistant Custom Integration's diagnostics beyond the scaffold baseline, conforming to spec/ha/diagnostics. Creates or edits diagnostics.py with asyncgetconfigentrydiagnostics (and optional asyncgetdevicediagnostics) returning structured dicts, and routes every…

nolte/claude-home-assistant · 199 tokens

System Healthcheck

Run a lightweight, non-destructive system healthcheck — disk usage, memory, CPU load, process count, hostname — and store a structured snapshot in session memory. Use when the user says "run a healthcheck", "check system health", "how's the server doing", "is everything ok", or asks for a status/diagnostics snapshot…

navig-run/core · 93 tokens

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

sagebynature/team-nexus · 21 tokens

mcp-control-plane: Skill for Codex