CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

verification-loop

A comprehensive verification system for Claude Code sessions.

2.42x
Quality

53%

Does it follow best practices?

Impact

92%

2.42x

Average score across 3 eval scenarios

SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./.agents/skills/verification-loop/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

75%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 well-structured, actionable verification playbook: six sequenced phases with concrete commands, explicit stop-on-failure checkpoints, and a clear output report format. Its weaknesses are modest — some padding in the Continuous Mode/Hooks sections, no enforcement command for the coverage target, and absent gating for the later phases.

DimensionReasoningScore

Conciseness

The body is dominated by lean, executable bash blocks with no explanations of concepts Claude already knows, so it mostly respects the token budget. Minor trimmable padding exists in the "Continuous Mode" block ("Set a mental checkpoint", the undefined "Run: /verify") and the "Integration with Hooks" section, keeping it just below the leanest anchor.

4 / 5

Actionability

Each phase supplies concrete, mostly copy-paste-ready commands ("npm run build 2>&1 | tail -20", "npx tsc --noEmit", "ruff check .", "git diff --stat") plus a concrete report template. Minor gaps — the 80% coverage target is only a comment with no enforcement command, and the lint/security phases give no instruction on what to do with findings — keep it below fully executable.

4 / 5

Workflow Clarity

The six phases are clearly sequenced and include an explicit checkpoint ("If build fails, STOP and fix before continuing", "Fix critical ones before continuing") plus a PASS/FAIL report format. Phases 3, 5, and 6 lack explicit gating or feedback loops, which is the minor validation gap that distinguishes this from the top anchor.

4 / 5

Progressive Disclosure

The skill is a single self-contained file with well-organized, clearly headed sections, and no content that clearly belongs in separate files (no references/ or scripts/ bundle exists). At ~120 lines it exceeds the under-50-line simple-skill exception and offers no reference signaling, so it earns good structure rather than the top score.

4 / 5

Total

16

/

20

Passed

Description

11%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 is a vague, buzzword-heavy one-liner that names neither the concrete verification actions (build, type check, lint, test, security scan, diff review) nor any usage triggers. It would rarely be selected for the right moment and is nearly indistinguishable from a generic quality-assistance skill.

Suggestions

List the concrete capabilities in the description, e.g. "Runs build verification, type checks, linting, tests with coverage, secret scanning, and diff review, then produces a PASS/FAIL verification report"

Add an explicit trigger clause, e.g. "Use when the user asks to verify changes, run quality gates, or check readiness before creating a PR"

Remove the buzzword "comprehensive" and name the scope (TypeScript/Python projects) so the skill is distinguishable from generic review/testing skills

DimensionReasoningScore

Specificity

The description states only "A comprehensive verification system for Claude Code sessions" — no concrete actions are named (no mention of builds, type checks, tests, or security scanning), and "comprehensive" is a buzzword over-claim. This matches the anchor for entirely vague, pure abstract language.

1 / 5

Completeness

The "what" is extremely vague (a "verification system" for what, exactly?) and the "when" is entirely absent — no "Use when..." clause or equivalent trigger guidance appears in the description (it only exists in the body). Both halves are effectively missing or contentless, matching the lowest anchor.

1 / 5

Trigger Term Quality

"verification" is one generic domain keyword, but the description lacks the natural phrases users would actually say such as "before creating a PR", "run the tests", or "check my code". It sits between having no relevant keywords and having some relevant keyword coverage with common variations missing.

2 / 5

Distinctiveness Conflict Risk

"A comprehensive verification system for Claude Code sessions" is very broad and would overlap with code-review, testing, and quality-gate skills, creating high trigger-conflict risk. It is at least scoped to verification, so it does not fall to entirely generic.

2 / 5

Total

6

/

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

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

15

/

16

Passed

Repository

saski/arnesto

Path

.agents/skills/verification-loop/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.