CtrlK
BlogDocsLog inGet started
Tessl Logo

verify

Prove a RuView result is real — run the deterministic SHA-256 proof and the witness bundle (ADR-028), and lint any claim for MEASURED-vs-CLAIMED honesty.

59

Quality

68%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./harness/ruview/.claude/skills/verify/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

78%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is a lean, well-structured verification workflow with concrete commands and strong per-section validation checkpoints; it falls just short of top marks on conciseness, actionability, and workflow clarity due to minor editorial padding, a few implicit execution details, and parallel rather than explicitly ordered sections.

DimensionReasoningScore

Conciseness

Tight sections with concrete commands and no concept over-explanation; only mild editorial padding ('The prove everything skill. Nothing ships as validated without this.') keeps it just below the lean/efficient 5 anchor.

4 / 5

Actionability

Concrete executable commands throughout ('ruview_verify', 'verify.py --generate-hash', 'bash scripts/generate-witness-bundle.sh', 'ruview_claim_check {text}') with real expected outputs; minor gaps (e.g., working directory, full flag set) keep it at 'mostly executable' rather than fully copy-paste ready.

4 / 5

Workflow Clarity

Each section carries an explicit validation checkpoint ('Must print VERDICT: PASS', 'must be 7/7 PASS') and a feedback loop (hash changed → regenerate → re-verify), but the four sections are parallel rather than a single ordered sequence, so checkpoints are mostly present with a minor ordering gap.

4 / 5

Progressive Disclosure

A single <50-line file organized into four well-signaled sections with no nested references; no bundle files exist so the simple-skill exception applies, allowing a 5 for well-organized sections alone.

5 / 5

Total

17

/

20

Passed

Description

58%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description conveys a specific, distinctive verification niche with several concrete actions, but omits an explicit 'Use when' trigger clause and leans on project jargon over natural user phrasing, which caps completeness and trigger-term quality.

Suggestions

Add an explicit 'Use when...' clause (e.g., 'Use when validating a RuView release, attesting a result, or quoting accuracy figures') to lift completeness above 3.

Soften jargon with natural synonyms users would actually say (e.g., 'verify', 'reproduce', 'check accuracy claims') to improve trigger-term coverage.

Consider naming the broader trigger surface (reports, PR bodies, model cards) inline since claim-checking already targets them.

DimensionReasoningScore

Specificity

Quotes 'run the deterministic SHA-256 proof and the witness bundle (ADR-028), and lint any claim for MEASURED-vs-CLAIMED honesty' — several concrete actions (proof, witness bundle, claim linting); not a comprehensive 5 because coverage of the verify domain is partial and 'RuView result' is the only domain named.

4 / 5

Completeness

Clear 'what' (run proof, witness bundle, lint claims) but no explicit 'Use when...' trigger clause; 'when' is only weakly implied by 'Prove a RuView result is real'. Per the missing-trigger-clause guideline, completeness is capped at 3.

3 / 5

Trigger Term Quality

Natural-ish phrase 'Prove a RuView result is real' is present, but surrounding terms ('witness bundle', 'ADR-028', 'MEASURED-vs-CLAIMED honesty') are domain jargon rather than common user phrasings, and synonyms are missing — matches the 'some relevant keywords but missing common variations' anchor.

3 / 5

Distinctiveness Conflict Risk

RuView-specific niche with distinct anchors (SHA-256 proof, ADR-028, MEASURED-vs-CLAIMED) gives a clear niche, but the generic verbs 'prove'/'verify' create minor overlap risk with other validation skills rather than the minimal conflict of a 5.

4 / 5

Total

14

/

20

Passed

Validation

93%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

referenced_paths_exist

Referenced path issues: 1 missing

Warning

Total

15

/

16

Passed

Repository
ruvnet/RuView
Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.