Skip to content
Agentmart
Submit a tool

verify

Run a distinct coordinator verification of a worker result.

Skill by douglasjarquin

IndexedMIT
Write a review

Chalk Labs

Need help deploying verify?

Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.

Book a call

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/douglasjarquin-verify-1fc574

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Run a distinct coordinator verification of a worker result. Use when a worker reports a candidate or findings, before treating that report as done.

Use cases: documents

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md and code. Checked Nov 1, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/verify

What's inside

Skill name
verify
Description
Run a distinct coordinator verification of a worker result. Use when a worker reports a candidate or findings, before treating that report as done.
License
MIT
Instructions
46 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

# Verify

## When to use it

A worker wrote `/workspace/sum/tasks/<id>/report.md`.
Use this before delivery and before telling the user the task is done.
Idle is not done.

## Required inputs and access

The task brief, the report, and the actual diff or findings.
The repository's own verification commands.
A checkout of the candidate that is not the worker's still-writing session.
Do not call a Cursor Cloud Agent.

## Sequence of work

1. Read the brief, the report, and the outstanding questions.
2. Confirm the checkout, branch, candidate, and actual diff.
3. Run the repository's own checks yourself against that candidate.
   This is a distinct run from the worker's.
   Reading the worker's logs is not that run.
   If `action-log.md` shows two `done` lines for the same intent and target, name that in `verification.md`.
   A duplicate does not pass a failing check.
4. Write `/workspace/sum/tasks/<id>/verification.md` with the commands, exit results, and what you actually saw.
5. If checks fail, send a bounded in-scope repair to the same worker against the task id; never ask a worker to add or reword code comments, other than a directive the code cannot work without, and never ask a worker to add, expand, rewrite or delete tests unless the user explicitly asked for that change.
   Repairs outside the approved brief are `expansion` and consume the allowance; stop after two of those and bring the budget question to the user.

## How to validate the result

`verification.md` names commands you ran, not commands the worker claimed.
The candidate in that file matches the report's candidate.
A screenshot or passing suite is not a substitute for the run the project itself defines.

## What to return

Pass, fail, or inconclusive, with the verification path.
A fail or inconclusive result stays visible.
Do not relabel it as done.

## What requires approval

Waiving missing evidence or a failing required check needs the user's explicit decision.
Do not invent that decision from the worker's report.

From douglasjarquin/sum, MIT.

More from douglasjarquin/sum

Signals

Stars
0
Forks
0
Open issues
2
Last commit (this folder)
Oct 8, 2026 (24 days ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from douglasjarquin

Done for you

Rather not set it up yourself?

Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.

See how we build AI agents →Read the case studies

Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.

launches shipped
50+
launches shipped
media placements
500+
media placements
founders on every account
2
founders on every account

Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block

Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.

Get verify running in your business

Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.

Goes to the Chalk Labs founders, not the newsletter. Privacy

Need help deploying verify?

CallGet help
Cite this page

verify skill. Agent Mart. https://agentmart.dev/skills/douglasjarquin-verify-1fc574 (last updated 2026-10-24)