What types of verification does it handle?

It covers unit tests, linters, build processes, bug fix validation, and general requirement checklists across any development environment.

Can it skip verification for small changes?

No, the core principle is 'no shortcuts.' Even minor changes must be verified with fresh evidence to prevent unexpected side effects.

Why is this skill necessary?

It prevents 'hallucinated' success by requiring Claude to run actual commands and read output before claiming a task is done, maintaining project integrity.

How does it handle delegated agent tasks?

It requires checking the VCS diff and verifying changes independently rather than trusting an agent's success report at face value.

Verification Before Completion

Name: Verification Before Completion
Author: clsandoval

byclsandoval

0•

Security & Testing

Enforces a strict evidence-based protocol for verifying code changes and task completion before making success claims.

This skill instills a rigorous 'Evidence Before Assertions' policy for Claude Code, preventing premature claims of success or completion. It mandates that every assertion—whether it's a passing test suite, a clean linting report, or a successful build—must be backed by fresh, verified command output. By eliminating assumptions and 'should-be' logic, this skill ensures higher code quality, maintains trust between the AI and developers, and prevents the introduction of regressions or incomplete features into the codebase.

Key Features

01Anti-Rationalization Guardrails: Identifies and stops the use of non-committal language like 'should' or 'probably'.

02Regression Protection: Mandates red-green cycles for bug fixes to ensure true resolution.

03Agent Output Validation: Independently verifies delegated tasks instead of trusting status reports.

04Evidence-Based Reporting: Replaces vague assurances with concrete CLI output and exit codes.

05Strict Gate Function: Blocks completion claims until verification commands are executed and confirmed.

060 GitHub stars

Use Cases

01Verifying a complex bug fix actually resolves the issue without causing regressions.

02Ensuring CI/CD pipelines will pass before committing or opening a Pull Request.

03Validating that all project requirements are met through a line-by-line checklist.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add clsandoval/monorepo verification-before-completion

For use in Claude.ai and ChatGPT

Key Features

01Anti-Rationalization Guardrails: Identifies and stops the use of non-committal language like 'should' or 'probably'.

02Regression Protection: Mandates red-green cycles for bug fixes to ensure true resolution.

03Agent Output Validation: Independently verifies delegated tasks instead of trusting status reports.

04Evidence-Based Reporting: Replaces vague assurances with concrete CLI output and exit codes.

05Strict Gate Function: Blocks completion claims until verification commands are executed and confirmed.

060 GitHub stars

Use Cases

01Verifying a complex bug fix actually resolves the issue without causing regressions.

02Ensuring CI/CD pipelines will pass before committing or opening a Pull Request.

03Validating that all project requirements are met through a line-by-line checklist.