Tessl • The management layer for AI agent skills

SkillPublic

validation-gates

The checks a change passes before it is done or merged: secrets, the project's tests, lint and build, size, new dependencies, the browser for UI, review, regression and a smoke test. Use when deciding which checks a change needs, or whether work is ready to merge or deploy.

Quality

88%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

85%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.

An efficient, tightly sequenced validation checklist with explicit feedback loops and mostly executable guidance. Its one structural defect is the dangling REFERENCE.md link in gate 4: the referenced commands for dependency vetting (maintained, licence, duplication) are unavailable, which weakens both actionability and progressive disclosure.

Suggestions

Create the missing REFERENCE.md (or remove the link) — gate 4 says "[REFERENCE.md](REFERENCE.md) has the commands" but the file is not in the bundle, so the licence/maintained/duplicate-check commands it promises cannot be reached.

Inline the dependency-vetting commands (e.g., `npm view <pkg> license`, `npm view <pkg> maintainers`, checking the existing manifest for duplicates) so gate 4 is self-contained if no separate reference file is intended.

Consider a one-line note on how to locate the task runner's skill commands for gate 1's "Tests, lint, build" step when `.opencastle/project.instructions.md` lacks them.

Dimension	Reasoning	Score
Conciseness	The 13-line body is lean and assumes Claude's competence — every token instructs ("Over 500 changed lines, 10 files or 2 projects: stop…", "Lint with auto-fix first") with zero padding or explanation of concepts Claude already knows.	5 / 5
Actionability	Mostly concrete: copy-paste commands like `gitleaks git --redact --log-opts="main..HEAD"` and `npm audit --audit-level=high`, explicit secret patterns (`AKIA…`, `sk-…`, `ghp_…`), and hard thresholds. Minor gaps remain: gate 4's maintained/licence/duplicate checks point to a REFERENCE.md that does not exist in the bundle, so that sub-guidance is not actually reachable.	4 / 5
Workflow Clarity	The eight gates are an explicit ordered sequence ("Run them in this order") with a built-in feedback loop ("when one fails, fix it and start again") — this skill itself is a validation workflow with checkpoints at every step.	5 / 5
Progressive Disclosure	The body correctly defers detail to a separate file ("[REFERENCE.md](REFERENCE.md) has the commands") and to named skills (browser-testing, fast-review, panel-majority-vote), but the bundle contains no references/, scripts/, or assets/ directories and no REFERENCE.md anywhere — the one file reference it makes is broken, so navigation fails at exactly the point of disclosure.	3 / 5
	Total	17 / 20 Passed

Description

88%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: it enumerates the concrete checks it covers, states its merge/deploy-readiness niche, and includes an explicit "Use when…" trigger clause in third person. Only minor gains available from adding synonyms like CI or quality-gate terminology.

Dimension	Reasoning	Score
Specificity	The description lists multiple concrete checks — "secrets, the project's tests, lint and build, size, new dependencies, the browser for UI, review, regression and a smoke test" — giving comprehensive, specific coverage of the change-validation domain rather than vague language.	5 / 5
Completeness	Both parts are explicit: the what ("The checks a change passes before it is done or merged: …") and the when ("Use when deciding which checks a change needs, or whether work is ready to merge or deploy") with concrete trigger phrases.	5 / 5
Trigger Term Quality	Natural user phrases like "checks", "lint and build", "review", "ready to merge or deploy" are present and would be said by users needing this skill. Coverage is good but misses a few common synonyms such as "CI", "quality gates", or "pre-merge checks".	4 / 5
Distinctiveness Conflict Risk	The merge-readiness gate niche is clear and mostly distinct, with phrases like "before it is done or merged" and "ready to merge or deploy". There is minor overlap risk with closely related review/testing skills (e.g., code-review, security-review) whose scope also touches "review" and "tests".	4 / 5
	Total	18 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
relative_links	Relative link issues: 1 missing	Warning

	Total	15 / 16 Passed

Repository: monkilabs/opencastle
Path: src/orchestrator/skills/validation-gates/SKILL.md
Revision: 296c05f

Reviewed: 1 month 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.

validation-gates

Table of Contents

Is this your skill?