Skills juliusbrussee caveman Caveman
verify-and-stop
- Use it when:
- A task only asks to check that existing work is done, such as a completion check, a focused gate run or last-mile proof.
- It does:
- Turns the acceptance conditions into the smallest set of checks that proves them, reusing results still current for the same code.
- You get:
- The commands run, their results and any unresolved risk, with each check marked pass, fail, unavailable or blocked.
- Won't:
- Edit product code unless you asked for fixes, or add polish, cleanup or unrelated tests once the criteria pass.
- Unlike surgical-patch:
- That one fixes a bug and then proves the fix; this one only proves.
- Runs:
- On its own, whenever the agent sees a task that fits; it's a short list of rules, not a step-by-step guide.
Skill 16 of 16 in the Caveman bundle, after migration.
TL;DR
One of Caveman's short work patterns, for tasks that ask for proof rather than changes. The agent translates the acceptance conditions into the smallest set of checks that would prove them. It reuses results that are still current for the same repository state, runs focused checks before wider gates, and keeps pass, fail, unavailable and blocked apart exactly. It doesn't edit product code unless the request includes fixes, and once the criteria pass it adds no polish, cleanup or unrelated tests. It stops as soon as the proof is complete, and reports only the commands, their results and any risk left unresolved.
Summarizes the skill at 4bf8898.
npx skills add https://github.com/juliusbrussee/caveman --skill verify-and-stop skills.sh pageSource on GitHubAll skills from juliusbrussee/caveman
More weekly downloads than 91% of skills
Weekly values
| Week to | Installs |
|---|---|
| Sep 21 | 13,099 |
| Sep 28 | 10,879 |
| Oct 5 | 10,087 |
| Oct 12 | 8,241 |
| Oct 19 | 7,107 |
| Oct 26 | 7,560 |
| Nov 2 | 7,182 |
| Nov 9 | 6,850 |
More downloads than 89% of skills
More stars than 93% of repos
Newer than 93% of skills
| Category |
Find root causes, review code and test what you built. Test-first work, implementation from specs, git hygiene and scope control. Everything for building software: planning, coding, debugging, design, frameworks, cloud, agents and skills. |
|---|---|
| Type |
A multi-step process the agent follows: interview, debug, run a TDD loop, review. |
| Effect |
Adds no side effects of its own. It shapes how the agent thinks, writes or reviews. Guidelines and routers are always effect:none, even when the task they support edits code. |
| Needs |
Instructions only. Works in any agent that supports skills. |
| Publisher |
Everyone else. Unverified vendor claims default to community. |
| Topics |
SKILL.md
From juliusbrussee/caveman at 4bf8898. Checked daily, last changed Nov 6, 2026.
Frontmatter
name: verify-and-stop
description: Prove existing work meets acceptance conditions without expanding scope. Use for validation-only tasks, completion checks, focused gate runs, and last-mile proof.Verify and stop
Translate acceptance conditions into smallest sufficient proof set.
- Reuse still-current results with matching repository state.
- Run focused checks before wider gates.
- Distinguish pass, fail, unavailable, and blocked exactly.
- Do not edit product code unless verification request includes fixes.
- Do not add polish, cleanup, or unrelated tests after criteria pass.
Stop immediately when acceptance proof is complete. Report commands, results, and unresolved risk only.