agentleFS
Sign inStart free
komlukkomluk/scaffolding/skills/verification-before-completion/SKILL.md

verification-before-completion

Skill16 starsChanged 5 days agoMITView on GitHub

Verify claims by executing, not asserting, before declaring work done. TRIGGER when: about to report a task complete, confirming a fix/build/deploy works, or delegating work to another agent or tool. SKIP: scoring research/plan quality pre-implementation (use quality-validation); deciding test coverage strategy (use testing-strategy).

189 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use chinese-code-review, chinese-commit-conventions and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • Leodorareluctant259Skill

    chinese-code-review

    中文代码审查规范——在保持专业严谨的同时，用符合国内团队文化的方式给出有效反馈

    Also in 8 projects that use this

  • Leodorareluctant259Skill

    chinese-commit-conventions

    中文 Git 提交规范 — 适配国内团队的 commit message 规范和 changelog 自动化

    Also in 8 projects that use this

  • obraSkill

    diagnosing-superpowers

    Use when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.

    Also in 20 projects that use this

  • jh941213Skill

    docs-ci

    문서 CI(링크·OpenAPI lint·breaking 게이트·mermaid 검증)나 문서 드리프트 감지(docs.yaml 신선도 검사)를 설치하거나 돌릴 때 사용. 문서 내용 작성은 docs-architecture·docs-interfaces·docs-manuals.

    Also in 3 projects that use this

  • jh941213Skill

    docs-interfaces

    API·이벤트 인터페이스 문서(OpenAPI·swagger·AsyncAPI 명세, 엔드포인트 레퍼런스, API 구성도·시퀀스, API CHANGELOG)를 만들거나 코드와 맞출 때 사용. 시스템 구성도는 docs-architecture, API 설계 원칙은 api-design-principles.

    Also in 3 projects that use this

  • +3 more from Leodorareluctant259
  • +4 more from jh941213

The file

We split this skill into the rules it gives an agent

1 rule. 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 177 edited versions we read keep these.

  • 1

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Do NOT apply this skill for:

    line 24

“Common” is measured: kept by at least 6 in 10 of the 177 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Verification Before Completion Skill

Purpose

A claim of "done" is only as good as the check that produced it. This skill encodes the rule that every completion claim, and every fact a claim depends on, must be backed by something actually executed and observed — not assumed, remembered, or guessed — before it is reported.

When to Apply

Apply this skill when:

  • About to report a task, fix, or feature as complete
  • Confirming that a build, test, deploy, or query actually succeeded
  • Stating a fact about a file, branch, or config that the response depends on
  • Delegating a task to another agent or tool

Do NOT apply this skill for:

  • Scoring a ResearchPack or Implementation Plan against quality gates before work starts — use quality-validation
  • Deciding what and how much to test — use testing-strategy

Verification Rules

RuleRequirement
Prove before claimingRun the actual command that demonstrates the outcome (test suite, build, the real query) and read its output. Exit code 0 alone is not proof of correctness — check the output content.
Verify against the authoritative sourceConfirm facts (file existence, path, branch content, config value) against the actual target (e.g. main, prod). A stale checkout, cached artifact, or side branch is not evidence about the target.
One failure = investigateAfter a failed attempt, diagnose the actual cause before trying again. Do not retry the same action with a blind variation.
Two guesses = stopIf a second consecutive guess is about to be made without new evidence, stop. Search the codebase for an existing pattern (grep) instead of guessing again.
Check capability before delegatingConfirm the executor (agent or tool) actually has the access/tools the task requires before handing it off.
Absence of error is not successNot seeing an error is not the same as confirming success. Only a positive, observed result counts as verification.
State the verificationReport what was verified and how (command run, output observed) so the claim is auditable by someone else.

Decision: Guess vs. Investigate vs. Search

SituationAction
First attempt failsRead the error, diagnose the actual cause
About to try a second unverified variationStop — grep the repo/docs for the existing pattern instead
Uncertain which transport/config/convention appliesSearch for how the codebase already does it before picking one
A fact is needed about main/productionRe-check it on that actual target, not a local/stale copy
Delegating a taskConfirm the recipient's tool/access list covers the task first

Anti-Patterns

Anti-PatternProblemInstead
Guessing a transport/config repeatedly (e.g. try HTTPS, then try SSH)Wastes turns on trial-and-error the repo already answersGrep the repo for the existing pattern already in use
Citing a file/path read from a stale checkout or different branchClaim doesn't hold on the actual target branchRe-read the fact from the actual target before stating it
Delegating to an agent/tool that lacks the needed capabilityFailure surfaces only after the delegate returnsConfirm required tools/access before handoff
Declaring "done" because no error appearedSilent failures pass unnoticedRun a command that positively demonstrates the result and read its output
Retrying an identical failed actionSame failure repeats, no progressDiagnose the cause before the next attempt

Quality Checklist

  • The command/query that proves completion was actually run, and its output read (not just its exit code)
  • Facts about files/branches/configs were checked against the authoritative target, not a stale or local copy
  • No two consecutive unverified guesses were made — the first failure triggered investigation, not a retry
  • The delegate's toolset/access was confirmed sufficient before handoff
  • The completion claim states what was verified and how (command + observed output)

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free