Tessl • The package manager for agent skills

An exemplary instruction-only skill body: lean, fully executable, and safely sequenced with explicit validation gates and an unambiguous output contract. The safety boundary (no npm publish / docker push / proof-log edits) and UNKNOWN handling show careful design.

Dimension	Reasoning	Score
Conciseness	Every line is operative — argument resolution rules, exact commands, gating conditions, an output template — with no concept explanations or padding. The single explanatory sentence about $ARGUMENTS substitution earns its place by making safe argument handling unambiguous.	5 / 5
Actionability	Fully executable guidance throughout: exact commands (`npm test`, `node src/cli.mjs verify "<path>"`, `npm run package:check`, `git rev-parse HEAD`, `git status --short`), a default path for empty arguments, quoting/safety rules for unsafe values, and a copy-paste-ready report template covering the common cases.	5 / 5
Workflow Clarity	Six numbered steps with explicit validation gates ("Run `npm run package:check` only when tests and candidate verification pass"), an evidence-log comparison checkpoint, and a PASS/FAIL/UNKNOWN reporting scheme that mandates naming any check that did not run. The workflow is itself a validation loop with explicit error-reporting, satisfying the anchor-5 pattern.	5 / 5
Progressive Disclosure	No bundle files exist and the body references only runtime project files (ISSUE.md, CLAUDE.md, evidence/PROOF-LOG.md) that are verification targets, not skill content to offload. At ~29 body lines with clear linear organization (preamble, procedure, limits, output template), it fits the simple-skill exception for a top score.	5 / 5
	Total	20 / 20 Passed

Description

73%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.

A tight, correctly formatted trigger description with an explicit Use-when clause and near-zero conflict risk. Its main weakness is thin capability coverage — one composite action instead of the concrete checks the skill actually performs — plus omission of the natural "verify" trigger word.

Suggestions

List the concrete checks performed, e.g. "Runs tests, verifies the candidate contract, does a package dry run, and compares git state against the evidence log."

Include the natural verb "verify" (and possibly "release readiness") so the description matches how a user would phrase the request.

Differentiate the when-clause from the what-clause with distinct trigger contexts (e.g. "Use when asked to verify or sign off on a Proofpack release candidate").

Dimension	Reasoning	Score
Specificity	Names a precise domain ("Proofpack release candidate" with "complete local evidence before packaging") but describes only a single composite checking action rather than the several concrete operations the skill performs (run tests, verify candidate contract, package dry run, compare git state).	3 / 5
Completeness	Both parts are present: a clear what ("checking whether a Proofpack release candidate has complete local evidence before packaging") and an explicit "Use when" clause. The when restates the what rather than offering distinct trigger contexts, so it falls short of the fully explicit anchor 5.	4 / 5
Trigger Term Quality	"checking", "release candidate", "evidence", and "packaging" are natural phrases a Proofpack user would say, giving good domain keyword coverage. It misses a few natural terms, notably "verify" — the skill's own name and the most likely user verb.	4 / 5
Distinctiveness Conflict Risk	"Proofpack release candidate" pins a clear niche with distinct triggers; no other plausible skill overlaps with verifying this specific project's release evidence, so conflict risk is minimal.	5 / 5
	Total	16 / 20 Passed

Validation

87%

Warnings & errors only

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

Validation — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: FlorianBruniaux/claude-code-ultimate-guide
Path: examples/learning-project/.claude/skills/verify-release/SKILL.md
Revision: f499217

Reviewed: 17 days ago

Discovery Implementation Validation

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.

verify-release

Table of Contents

Is this your skill?