skills.rest
marcocpt
marcocptCOMMUNITY
Explore all skills by this author →
GitHub

verification-before-completion

Enforces running verification commands before claiming tasks are complete or tests pass.

Updated Jun 11, 2026
Search Tags:#code quality#testing#verification#test validation#evidence-based#build verification#completion claims
One-click install
npx skills add https://github.com/marcocpt/trae_skills --skill verification-before-completion-marcocpt
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: verification-before-completion
Source: https://github.com/marcocpt/trae_skills/tree/main/verification-before-completion
Command: npx skills add https://github.com/marcocpt/trae_skills --skill verification-before-completion-marcocpt

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? AI agents and developers often claim work is done, tests pass, or bugs are fixed without actually running verification commands, leading to broken code, missed requirements, and lost trust. This Skill enforces an evidence-first discipline: no completion claim without fresh verification output. ## Core Features & Use Cases - Verification Gate: A mandatory checklist requiring you to identify the proving command, run it fully, read the output and exit codes, and only then state conclusions with evidence. - Failure Pattern Catalog: Tables mapping common claims (tests pass, build succeeds, bug fixed, requirements met) to the exact evidence required versus insufficient shortcuts. - Rationalization Blockers: Explicit counters for excuses like "it should work", "the linter passed", or "the agent reported success". - Use Case: Before committing code or creating a PR, the Skill forces you to re-run the test suite, confirm zero failures in the output, and only then report completion—preventing shipping undefined functions or unverified fixes. ## Quick Start Before claiming any task is complete, run the relevant verification command and report the actual output as evidence for your conclusion.

Frequently Asked Questions about verification-before-completion

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I verify tests actually pass before claiming completion?▼

Run the full test command fresh in the current session, read the complete output, and confirm zero failures before stating tests pass. Previous runs, partial checks, or assumptions like "it should pass" do not count as evidence.

What counts as valid evidence that a bug is fixed?▼

Valid evidence means re-testing the original symptom and observing it pass after the fix. Changing the code and assuming the fix works is insufficient—you must run the failing scenario and confirm the corrected behavior.

Does a passing linter mean the build succeeds?▼

No, a linter passing does not prove compilation or build success. Linters check style and static issues but do not compile code; you must run the actual build command and confirm exit code 0.

How do I verify a regression test actually works?▼

Use the red-green cycle: write the test, confirm it passes with the fix, revert the fix and confirm the test fails, then restore the fix and confirm it passes again. A test that only ever passed once is unverified.

Can I trust an AI agent's report that a task succeeded?▼

No, agent success reports require independent verification. Inspect the VCS diff to confirm actual changes were made and validate the results yourself before reporting the delegated task as complete.

When should verification run in a development workflow?▼

Run verification before any success claim, expression of satisfaction, commit, PR creation, task completion marking, or handoff to another agent. Any wording implying success without fresh evidence violates the rule.

Related Skills

Explore More Software Engineering Skills→
zengbaochengzengbaocheng

test-driven-development

Enforce a test-driven development workflow with failing tests before coding.

Community
Intermediate
pnewsampnewsam
1

deliver-work

Delivers work candidates to requested endpoints such as commits, pull requests, and merges.

Community
Advanced
519lab519lab
2

requesting-code-review

Run automated code review with security scans and static analysis before commit.

Official
Advanced
mahg-esmahg-es
3

cicd-quality

Design and enforce CI/CD quality gates for code, tests, security, build, E2E, and performance.

Official
Advanced