Skip to content

Maintain Verification Skill

Keeps a project's verification skill and feature/agent access maps aligned with source code and live evidence.

Stars
110,078
Licence
MIT
Files
3 files

What it does

This is a maintenance playbook for an AI agent working on the Paperclip codebase. When asked, the agent audits an existing project-local verification skill and its feature and agent-access maps against the current source code, fixes proven outdated references or recipes, and reports what is covered, what is changed, and what gaps remain. It tells the agent to read source routes and permissions, run the repo's map structure check, and drive the running app through the existing verification workflow, while keeping product code and permission rules off-limits for edits.

Great for

  • Paperclip maintainers keeping their verification skill and product maps current after code changes
  • Developers auditing whether every business operation has an API and a permitted agent path
  • Teams that want honest coverage reports separating source review from live proof

Skip it if

  • Ordinary end users with no existing verification skill, since it assumes one already exists and points to a separate creation skill otherwise
  • Anyone expecting it to change product APIs or permissions, which it explicitly forbids during a maintenance run
  • Projects outside the Paperclip layout where map locations differ, since it only defines conventions for Paperclip and defers to existing locations elsewhere

Try asking

  • “/maintain-verification-skill”
  • “Audit the verify skill for the agent access map after the recent routes change”
  • “Reconcile feature and agent access coverage for the issues area and report remaining gaps”
  • “Run a full verification maintenance pass and separate source-only findings from live results”
Step-by-step guide: how to use Maintain Verification SkillInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 3 original files, the MIT licence and an attribution note.

    maintain-verification-skill.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip maintain-verification-skill.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip maintain-verification-skill.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Name a specific area for a targeted audit, since a targeted run is reported only for its named areas and not as a full pass
  • Expect the live pass to start and drive a real local app, so check that no other instance is in use before running
  • Review proposed map and skill corrections in a PR, since the skill keeps evidence there rather than in a shared ledger

What's inside

  • agents/openai.yaml
  • NOTICE.md
  • SKILL.md

Redistributed unmodified from paperclipai/paperclip under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.