skills.rest
taewook486
taewook486COMMUNITY
Explore all skills by this author →
WebsiteGitHub

moai-workflow-docs-claim-check

Validates documentation claims against supplied evidence with labeled atomic assessments.

Updated Mar 22, 2026
Search Tags:#validation#readme#release notes#evidence#read-only#documentation audit#claim verification
One-click install
npx skills add https://github.com/taewook486/Masters_degree --skill moai-workflow-docs-claim-check-taewook486
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: moai-workflow-docs-claim-check
Source: https://github.com/taewook486/Masters_degree/tree/main/.claude/skills/moai-workflow-docs-claim-check
Command: npx skills add https://github.com/taewook486/Masters_degree --skill moai-workflow-docs-claim-check-taewook486

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Public-facing documents like READMEs and release notes often contain claims that are outdated, unsupported, or unverifiable, and manually auditing them claim-by-claim is slow and inconsistent. This Skill performs a read-only audit that splits every statement into atomic claims and labels each one against the evidence you supply. ## Core Features & Use Cases - Atomic Claim Triage: Decomposes composite sentences into single-assertion claims so each receives exactly one label, and excludes subjective statements from labeling. - Ordered Label Decision Tree: Assigns one of four labels (needs-human, stale-suspected, verified, unsupported) via a strict gate order, with exactly one reason for every unsupported claim. - Hard Read-Only Boundaries: Runs no commands, writes no fixes, and performs no code or security review; claims requiring execution are labeled needs-human with the exact command named. - Use Case: Before publishing a release note, supply the note plus your manifest, CI logs, and artifact listings, and receive a three-section report showing which claims are verified, which are stale, and which lack evidence. ## Quick Start Audit the attached README against the supplied pyproject.toml and CI logs and report which claims are verified, stale, or unsupported.

Frequently Asked Questions about moai-workflow-docs-claim-check

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

FAQPage Schema
How do I check whether README claims are supported by evidence?▼

Supply the README plus evidence files such as manifests, CI logs, or artifact listings. The skill extracts every checkable statement, splits composite claims into atomic ones, and labels each as verified, stale-suspected, unsupported, or needs-human with named evidence anchors.

What labels does a documentation claim audit produce?▼

Each atomic claim receives exactly one of four labels: needs-human when settling it requires running something, stale-suspected for version or count drift, verified when supplied evidence directly supports it, and unsupported with one reason: missing-evidence, contradicted, or insufficient-coverage.

Can the claim check run tests or commands to verify a claim?▼

No. The skill is strictly read-only and never executes commands, builds, or tests. When a claim can only be settled by running something, it names the exact command and target file and labels the claim needs-human.

Does the claim check fix or rewrite unsupported documentation?▼

No. It produces findings only and never outputs patches, diffs, or rewritten passages. For each problem claim it describes what a maintainer would change and where, leaving the edit as a maintainer action.

What happens when no evidence is supplied for a documentation audit?▼

An empty evidence set is valid input and never produces a fabricated verified label. Every claim is labeled unsupported with missing-evidence or needs-human, and the report lists exactly what evidence the maintainer still needs to supply.

How are subjective statements like fast or intuitive handled?▼

Subjective statements cannot be checked against evidence, so they are excluded from labeling and listed under Input Scope Reviewed with a one-line reason. When a subjective adjective wraps a checkable core, the core is labeled and the adjective excluded.

Related Skills

Explore More Software Engineering Skills→
ManfronEnricoManfronEnrico

ml-audit-repo

Audit ML repositories by verifying claims across code, data, training, and metrics.

Community
Advanced
taewook486taewook486

moai-workflow-docs-claim-check

Validates documentation claims against supplied evidence with labeled atomic assessments.

Community
Intermediate
DozyboyDozyboy

hs:research

Gathers and verifies multi-source technical evidence into a ranked research report.

Community
Advanced
lth0lth0

verification

Run deterministic verification commands before asserting research or writing success.

Community
Intermediate