Home / Skills / verification-methodology

verification-methodology

Guides agents through pass/fail verification using explicit criteria, direct source-faithful evidence, reproducible checks, and structured verdicts before declaring work complete.

★ 88 GitHub Stars

86 Quality score

7 Files

About this Skill

Verify work against explicit criteria using direct, source-faithful evidence, reproducible checks, and clear verdicts. Use before declaring an artifact, implementation, or claim complete; do not use for exploratory research without pass/fail criteria.

Use cases

Assessing implementations against completion criteria
Fact-checking claims against primary sources
Verifying editorial or content changes
Reconciling evidence from multiple verification lanes
Producing structured verification verdicts

Features

Five-step verification protocol
Source-fidelity rules for direct evidence collection
Pass, fail, blocked, and not-applicable verdict states
Evidence and criteria assessment references
Defined 12-criterion Refine-to-Ship gate

Supported Agents

Not specified

Skill files

Other


                                evals/evals.json

JSON Other


                                README.md

Markdown Reference


                                references/criteria-assessment.md

Markdown Reference


                                references/evidence-standards.md

Markdown Reference


                                references/source-index.md

Markdown Reference


                                references/verdict-template.md

Markdown SKILL.md


                                SKILL.md

Markdown

Related Skills

Testing 96

agent-evals-and-observability

Provides framework- and vendor-neutral methods for evaluating AI agents, reviewing trajectories, analyzing regressions, setting release gates, and designing privacy-aware producti…

magnus919/agent-skills ★ 88

agent evaluation observability trajectory review regression analysis

Testing 96

playwright

Enables agents to author, run, debug, and triage Playwright end-to-end tests, integrate suites with CI, perform headless browsing and scraping, and check accessibility snapshots.

magnus919/agent-skills ★ 88

playwright e2e testing browser automation ci

Testing 96

qa-methodology

Provides software teams with guidance, templates, references, and Python scripts for test strategy, regression, risk-based testing, CI triage, quality gates, exploratory testing, …

magnus919/agent-skills ★ 88

quality assurance test strategy test automation regression testing

Testing 91

request-timing-audit

Audits iOS and macOS projects for permission requests made at launch or before meaningful user interaction, identifying likely App Store Guideline 5.1.1(ii) violations.