capcheck

A Go code checker that records which privileged actions a module and its dependencies can perform, such as starting processes, opening network connections, using cgo, or writing files.

In plain words
What is it for?
Use it in Go projects to create and check a capability baseline during dependency reviews or CI. It can list capabilities, compare them with a saved baseline, and report changes in text, JSON, or GitHub workflow format.
Why use it?
It detects when an update adds a new capability, even when no known security vulnerability has been reported. This helps catch supply-chain risk before accepting a dependency change.

Skill for Claude CodeCodex

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.

agentmods
npx agentmods add skills/git-pkgs/skills/capcheck
Any agent
npx skills add git-pkgs/skills --skill capcheck
Clone the repo
git clone --depth 1 https://github.com/git-pkgs/skills

Made for: Claude Code, Codex.

Per session 62 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 767 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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.

ModelPer sessionOnce invoked
Fable 5 $0.00062 $0.00767
Opus 5 $0.00031 $0.00383
Sonnet 5 $0.00012 $0.00153
Haiku 4.5 $0.00006 $0.00077

Measured yesterday against content hash bd5cc7cb9974, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

capcheck 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 yesterday.

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.

skills/capcheck/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.

Go Capability Drift

capcheck records the set of privileged operations reachable from a Go module (via google/capslock) into a lock file, then fails when that set grows. govulncheck reports CVEs; capcheck reports when a dependency gains the ability to do something it couldn't before, whether or not a CVE exists yet.

Go only. Run from the module root.

Commands

First-time setup (writes capcheck.json config and capcheck.lock.json baseline; commit both):

capcheck init ./...

Check against the baseline (default command, exit 1 on new capabilities):

capcheck ./...

Accept the current state after reviewing a reported change:

capcheck update ./...

Print current capabilities without a baseline:

capcheck list ./...

Flags (all commands):

  • -f text|json|github - output format; github emits workflow annotations
  • --strict - also fail on removed capabilities
  • --ignore CAP - ignore a capability (repeatable, stacks with config)
  • --granularity package|function - function is more precise but noisier
  • -C DIR - run as if in DIR
  • --baseline PATH - lock file path

Configuration

capcheck.json (all keys optional):

{
  "granularity": "package",
  "timeout": "5m",
  "goos": "linux",
  "goarch": "amd64",
  "ignore": ["FILES", "NETWORK", "REFLECT", "RUNTIME"]
}

Most projects ignore the noisy capabilities and watch for EXEC, CGO, ARBITRARY_EXECUTION, UNSAFE_POINTER, and MODIFY_SYSTEM_STATE. ignore matches hierarchically, so MODIFY_SYSTEM_STATE also covers MODIFY_SYSTEM_STATE/ENV.

Results depend on which stdlib files compile in, so pin goos/goarch to whatever CI runs on and keep it fixed; a lock file written on macOS will not match one written on Linux.

GitHub Action

- uses: git-pkgs/capcheck@v1
  with:
    packages: ./...

Runs check --format github and annotates the first line of user code in each new capability's call path.

Read the full file on GitHub · 88 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.

  1. yesterday First seen · 88 lines · 62 tokens per session scan A bd5cc7cb9974

Subscribe to this mod's changes

capcheck is a skill published in the GitHub repository git-pkgs/skills (1 stars, last pushed 12d ago), licensed MIT. It adds 62 tokens to every session and 767 once invoked, about $0.0003 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.

Related

Other skills, from other repositories

supply-chain-guard

Use before installing, updating, auditing, or executing dependencies, package-manager commands, project generators, CI actions/workflows, release jobs, IDE extensions, MCP servers, or AI-agent tools. Also use when investigating suspected compromise or advisories, debugging publish or release authentication, or…

ZarK/ai-supply-chain-guard · 71 tokens

managing-dependencies

Evaluates packages, manages dependencies, and addresses supply chain security. Use when adding npm/pip/cargo/bundler/go dependencies, auditing packages, reviewing lockfile changes, checking for vulnerabilities, comparing package alternatives, or assessing package trustworthiness.

andrew/managing-dependencies · 54 tokens

systematic-debugging

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

obra/superpowers · 21 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

babysit-pr

Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…

openai/codex · 114 tokens

imagegen

Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…

openai/codex · 113 tokens