A security and alignment layer for AI agents that monitors actions, checks sensitive operations, records significant work, and keeps sessions tied to the user's goals.

Use it to check the agent's security status, load goals, verify sensitive actions, log activity, and refresh goal context during long sessions.

It helps catch unsafe actions and work that has drifted away from the original request.

carapace — Skill for Claude Code and Codex by jdesl55

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent

npx skills add jdesl55/carapace-mcp --skill carapace

Clone the repo

git clone --depth 1 https://github.com/jdesl55/carapace-mcp

Made for: Claude Code, 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/jdesl55/carapace-mcp/carapace/github.svg)](https://agentmods.dev/skills/jdesl55/carapace-mcp/carapace)

Your own site

<a href="https://agentmods.dev/skills/jdesl55/carapace-mcp/carapace"><img src="https://agentmods.dev/badge/skills/jdesl55/carapace-mcp/carapace/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/jdesl55/carapace-mcp/carapace"><img src="https://agentmods.dev/badge/skills/jdesl55/carapace-mcp/carapace.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 0 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 885 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	$0.00000	$0.00885
Opus 5	$0.00000	$0.00443
Sonnet 5	$0.00000	$0.00177
Haiku 4.5	$0.00000	$0.00089

Measured 9d ago against content hash 79159d1e0ccf, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

carapace 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 9d ago.

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.

skill/carapace/SKILL.md · 73 lines

How it starts

The opening of the file, as written. The whole thing — 73 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Carapace Skill — Armor for your agent

You are operating with Carapace, a security and alignment layer that monitors your actions, verifies sensitive operations, and keeps you anchored to the user's goals.

Session Startup Protocol

At the start of every session, follow these steps in order:

Call carapace_status to confirm the server is running and check your security posture.
Call carapace_anchor to load the user's goals, priorities, and constraints. Treat the returned anchor block as your grounding truth for this session.
Read ~/.carapace/insights.md if it exists. These are learnings from your previous sessions. Incorporate them into your approach for this session.
Proceed with the user's requested tasks, following the operating rules below.

Operating Rules

Before any sensitive action: Call carapace_verify first. Do NOT proceed if the verdict is "block".
After any significant action: Call carapace_log to record what you did.
Every 15 minutes: Call carapace_anchor with a summary of your recent activity to refresh your goal context and check for drift.
Include the verification key: When executing a verified action, include the verification_key from the most recent carapace_verify response.
Respect the tier system:
- Tier 1 (highest risk): Financial transactions, sending messages, file deletion, account changes. Always verify first.
- Tier 2 (medium risk): Web browsing, package installs, API write calls. Verify when possible.
- Tier 3 (low risk): File reads, calendar checks, info lookups. Log when relevant.

Tool Reference

carapace_verify

When: BEFORE any sensitive action (Tier 1 mandatory, Tier 2 recommended) Input:

action_type: Category of the action (e.g., send_email, make_purchase, delete_file)
target: Who/what the action targets
description: What you intend to do
amount: Dollar amount if spending is involved (optional)
current_key: Your most recent verification key (optional)

Read the full file on GitHub · 73 lines

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.

9d ago First seen · 73 lines · 0 tokens per session scan A 79159d1e0ccf

Subscribe to this mod's changes

carapace is a skill published in the GitHub repository jdesl55/carapace-mcp (0 stars, last pushed 6mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 885 tokens. 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 2026-09-01.

Security

systematic-debugging

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

obra/superpowers · 21 tokens

local-ai-agents

Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…

microsoft/ai-agents-for-beginners · 200 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

insight-error-page

Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…

vercel/next.js · 83 tokens

next-cache-components-optimizer

Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…

vercel/next.js · 170 tokens

next-partial-prefetching-adoption

Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…

vercel/next.js · 103 tokens