Tessl • The management layer for AI agent skills

SkillPublic

verify

Use when about to claim work is complete, fixed, or passing - before committing, creating PRs, or moving to next task. Requires running verification commands and confirming output before making any success claims.

Quality

79%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/verify/SKILL.md

Quality

Content

81%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 highly actionable — concrete commands, a claim-to-evidence matrix, a gated workflow with explicit validation and failure branches, and ready-to-use output templates. Its main weakness is token efficiency: the Red Flags and Rationalization Prevention sections (and Iron Law vs. Rules) say the same things twice, adding ~30 lines of pure redundancy.

Suggestions

Merge 'Red Flags - STOP' and 'Rationalization Prevention' into a single excuse-vs-reality table — they list the same rationalizations ("Should work now", "Lint passed", "I only changed one line") almost verbatim.

Fold 'The Iron Law' into the 'Rules' section (or vice versa) since both restate 'evidence before claims, no exceptions'.

Trim the second (failure) output template and the 'Integration' agent list, or move them to a reference file, to tighten the body without losing the core gate workflow.

Dimension	Reasoning	Score
Conciseness	The 'Red Flags - STOP' list and the 'Rationalization Prevention' table repeat the same excuses nearly verbatim ("Should work now", "Lint passed, so build will too", "I only changed one line" appear in both), and 'The Iron Law' section overlaps the closing 'Rules' section. Not 2 because nothing explains concepts Claude already knows and the tone is punchy; not 4 because this duplication is genuine padding that could be cut.	3 / 5
Actionability	Provides copy-paste-ready commands for each operation (`npm test 2>&1`, `npm run build 2>&1`, `npm run lint 2>&1`, `npx tsc --noEmit 2>&1`), a claim-to-command Verification Matrix, and concrete output templates with expected counts and exit codes covering the common cases.	5 / 5
Workflow Clarity	The Gate Function is a clearly sequenced 5-step workflow (IDENTIFY, RUN, READ, VERIFY, claim) with an explicit validation checkpoint and failure branch ("If NO: State actual status with evidence", "DO NOT claim completion. Fix failures first"), plus a red-flags checklist. This matches the top anchor for explicit validation and error-recovery feedback.	5 / 5
Progressive Disclosure	Sections are well-organized and easy to navigate (When to Use, Gate Function, Verification Matrix, Operations, Output, Rules) and the skill appropriately needs no external reference files. Not 5 because at ~170 lines with visibly duplicated sections (Red Flags vs. Rationalization Prevention, Iron Law vs. Rules) the content could be consolidated or trimmed.	4 / 5
	Total	17 / 20 Passed

Description

70%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 strong description with an explicit 'Use when...' trigger clause covering natural completion-related phrases and a clear behavioral 'what'. Its main weakness is that the capability statement stays generic — it says verification commands must run but never names which verifications (test, build, lint, types) the skill performs.

Suggestions

Enumerate the concrete verification actions in the description, e.g. 'Runs tests, builds, lint, and type checks and requires reading full output and exit codes before any completion claim', to lift specificity.

Add common trigger synonyms such as 'done', 'finished', 'verify', or 'it works now' so the description fires on the natural phrasings users actually say.

State the scope boundary (e.g. 'NOT for research or exploration tasks') in the description itself to further reduce overlap with exploratory or review-type skills.

Dimension	Reasoning	Score
Specificity	The description names the domain (verification before completion claims) and states 1-2 required actions ("Requires running verification commands and confirming output"), but the actions remain generic — it never names tests, builds, lint, or type checks. It is not 2 because two concrete required actions are explicitly stated, and not 4 because it does not list several specific actions.	3 / 5
Completeness	Explicitly answers both: what ("Requires running verification commands and confirming output before making any success claims") and when ("Use when about to claim work is complete, fixed, or passing - before committing, creating PRs, or moving to next task"). Not 5 because the 'what' is a single general behavioral rule rather than a concrete list of capabilities.	4 / 5
Trigger Term Quality	Includes natural phrases users would actually say — "claim work is complete, fixed, or passing", "before committing, creating PRs", "moving to next task". Good coverage, but misses common synonyms such as "done", "verify", or "it works".	4 / 5
Distinctiveness Conflict Risk	The verification-before-completion niche with triggers like "before committing, creating PRs" is mostly distinct from other skills. Minor overlap risk with testing or code-review skills that also fire around commits and PRs.	4 / 5
	Total	15 / 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: fernandezbaptiste/specialist-agent
Path: skills/verify/SKILL.md
Revision: cd716d9

Reviewed: 8 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

Table of Contents

Is this your skill?