evaluation-domains

Evaluate source credibility using domain profiles with CRAAP axes and verification procedures.

Updated Aug 2, 2026
One-click install
npx skills add https://github.com/leonardoacosta/skills --skill evaluation-domains-leonardoacosta
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: evaluation-domains
Source: https://github.com/leonardoacosta/skills/tree/main/meta/skills/evaluation-domains
Command: npx skills add https://github.com/leonardoacosta/skills --skill evaluation-domains-leonardoacosta

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Evaluating whether a source, repository, or news item is trustworthy usually means each caller hardcodes its own ad-hoc criteria. This Skill provides a shared, data-shaped contract — domain profiles with fixed sections for credibility axes, verification procedures, verdict vocabulary, and memory homes — so any evaluation workflow can route to a consistent, reusable standard. ## Core Features & Use Cases - Domain Profile Contract: Every profile answers six fixed sections (primary-source definition, CRAAP axis weights, staleness horizon, verification procedure, verdict vocabulary, memory home), and adding a new domain requires only one conforming profile file. - Shared Vocabulary and Procedures: A central reference defines the five CRAAP credibility axes and three verification procedures (signal-bundle liveness, SIFT/lateral reading, evidence-audit) that profiles cite rather than restate. - Record-Keeping Doctrine: Six binding rules (R1-R6) govern how evaluation records are stored — append-only logs, progressive disclosure, staleness re-verification, and provenance chains. - Use Case: When vetting an external GitHub repo for adoption, route to the external-repos profile to apply code-over-README authority weighting, the duplicate-detection gate, and the Steal/Adapt/Monitor/Skip verdict vocabulary. ## Quick Start Use the evaluation-domains skill to vet this external repository and tell me which credibility profile and verification procedure apply.

Frequently Asked Questions about evaluation-domains

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

FAQPage Schema
How do I evaluate whether a source or domain is credible?

Route the source to its domain profile, which defines a primary-source hierarchy, weighted CRAAP axes (Currency, Relevance, Authority, Accuracy, Purpose), and a verification procedure. The profile produces a verdict from its defined vocabulary and records it in a memory home.

How do I add a new evaluation domain profile?

Author a new file under references/profiles answering all six contract sections: primary-source definition, credibility axes with weights, staleness horizon, verification procedure, verdict vocabulary, and memory home. No changes to the spec or consuming commands are required.

What verification procedures does source vetting support?

Three procedures are defined: signal-bundle liveness for judging repo or release health from multiple independent signals, SIFT/lateral reading for open-web and UGC sources, and evidence-audit for re-verifying every citation behind a drafted finding.

When should a recorded evaluation verdict be re-verified?

Any record older than the domain's staleness horizon — defaulting to 30 days — must be re-verified against its primary source before its verdict is reused. The horizon is an outer bound, not a recommended polling interval.

What are the limitations of single-indicator liveness checks?

Judging a project from one context-free indicator like star count or last commit date is explicitly banned. Liveness requires a bundle of independent signals such as commit recency, issue response latency, release cadence, and contributor trends.