CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

verify

This skill should be used when the user asks to "verify protocols", "check consistency before commit", "validate definitions", "run pre-commit checks", "verify soundness", or wants to ensure epistemic protocol quality. Invoke explicitly with /verify for pre-commit validation.

Quality

77%

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 ./.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.

A well-engineered instruction-only skill: an executable command with a defined output contract, clearly sequenced phases with error handling, and exemplary progressive disclosure that keeps checklists and criteria in verified one-level reference files with no duplication. The only improvements are minor: trim the duplicated output examples, give a concrete subagent invocation pattern, and close the fix-then-reverify loop.

DimensionReasoningScore

Conciseness

The body is efficient — tables for principles, severity mapping, and error handling; no explanations of concepts Claude already knows; references defer detail instead of inlining it. Not 5 because the Phase 4 output template is duplicated by two full 'Output Examples' sections that restate the same format and could be trimmed to one. Not 3: the padding is limited to that one redundancy.

4 / 5

Actionability

Concrete, executable guidance throughout: the exact command `node ${SKILL_DIR}/scripts/static-checks.js ${PROJECT_ROOT}`, the JSON output schema, severity synthesis rules, four explicit user options, and the commit annotation format. Not 5 because Phase 2's subagent spawning defers mechanics (how to sample the diff, how results are returned) without a concrete invocation pattern, leaving a minor gap.

4 / 5

Workflow Clarity

Five phases are clearly sequenced with an error-handling table covering script failure and subagent timeout (feedback for error recovery) and an explicit user decision gate. Not 5 because Phase 5's 'Fix critical issues' path never instructs re-running verification after fixes, so the validate→fix→re-validate loop is left implicit — a minor validation gap.

4 / 5

Progressive Disclosure

The body is an overview that appropriately splits detail into one-level-deep, well-signaled references (criteria.md, review-checklists.md, verification.md — all verified to exist and to contain exactly what the body claims) and script files, explicitly maintaining a single source of truth ('do not duplicate the templates here'). All referenced paths are real; co-change.md is a clearly-signaled sibling cross-link, not buried content, so navigation is easy.

5 / 5

Total

17

/

20

Passed

Description

66%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 excels at trigger guidance — explicit, natural, and rich with synonyms and an invocation cue — but it is entirely when-focused: it never states what the skill concretely does (static checks, expert review, severity-categorized findings). Its broadest trigger phrases also risk colliding with generic pre-commit/lint skills.

Suggestions

Add a leading capability statement naming the concrete mechanisms, e.g., 'Runs zero-context static checks and three parallel expert reviews (type theory, instruction design, Claude Code ecosystem) on protocol definitions, surfacing findings by severity.'

Narrow the generic triggers: replace 'run pre-commit checks' with protocol-scoped phrasing like 'run pre-commit checks on protocol definitions' to reduce overlap with generic lint/CI skills.

Consider mentioning what the verification covers (Definition blocks, prose surfaces, repository artifacts) so the 'what' matches the richness of the 'when'.

DimensionReasoningScore

Specificity

The description names the domain and quotes action phrases like "verify protocols", "validate definitions", and "run pre-commit checks", but these are all synonyms for a single verify action; the skill's actual capabilities (static checks, parallel expert subagent review, severity synthesis) are never stated. This matches the anchor 'Names domain and 1-2 concrete actions, but not comprehensive' rather than 4, which requires several distinct specific actions.

3 / 5

Completeness

The 'when' is explicit and thorough ("This skill should be used when the user asks to..." plus "Invoke explicitly with /verify for pre-commit validation"), and a 'what' exists ("ensure epistemic protocol quality"), but the 'what' is thin — it never says what the skill actually does. Fits anchor 4 ('both present; one could be more explicit') rather than 5, whose example states concrete capabilities alongside triggers.

4 / 5

Trigger Term Quality

Good natural-language coverage: "verify protocols", "check consistency before commit", "validate definitions", "run pre-commit checks", "verify soundness", plus the explicit "/verify" invocation cue, with verify/validate/check synonym variation. Not 5 because a few natural terms a user might say (e.g., "audit protocols", naming protocol files) are missing.

4 / 5

Distinctiveness Conflict Risk

The epistemic-protocol domain is a clear niche, but trigger phrases like "run pre-commit checks" and "check consistency before commit" would naturally be said for generic linting/CI/commit-review skills, creating real overlap risk. Fits 'Somewhat specific but could still overlap with similar skills' better than 4's 'minor overlap risk only'.

3 / 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.

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

allowed_tools_field

'allowed-tools' contains unusual tool name(s)

Warning

Total

15

/

16

Passed

Repository

jongwony/epistemic-protocols

Path

.claude/skills/verify/SKILL.md

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.