spec-validate

Validates a mainspec and its slices via 3-subagent consensus plus expert review, then *applies* impactful fixes directly to the spec files. No human-in-the-loop summary or approval — agent-first. Touches `specs/<feature>/.validated` as its final committed action.

Quality

66%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Critical

Do not install without reviewing

Fix and improve this skill with Tessl

tessl review fix ./skills/sdd/spec-validate/SKILL.md

Quality

Content

77%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 and its multi-phase workflow is sequenced with explicit checkpoints, validation, and crash-recovery handling. The main weaknesses are conciseness — a redundant Notes section and overlap between the completion protocol and the phase descriptions — and the absence of any progressive disclosure, with everything inlined in one long file.

Suggestions

Delete or trim the 'Notes' section, since every bullet restates guidance already given in the Invocation Contract or Phases; keep only any point not already covered.

De-duplicate the completion protocol against the Phase 1–3 descriptions so the ordered 7-step list and the phase bodies do not re-explain the same steps twice.

Move the full impactful/nitpick classification examples and the validation-log template into a reference file (e.g. references/classification.md) referenced one level deep, so SKILL.md stays an overview.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient task guidance, but the 'Notes' section restates points already made in the body ('Always 3 subagents', 'Model: opus', 'Foreground parallel execution', 'Consensus = confidence', 'All fixes applied without HITL') and the Invocation Contract completion protocol overlaps the Phase descriptions, adding padding that could be tightened without losing information.	3 / 5
Actionability	It provides fully executable, copy-paste-ready artifacts: concrete Task() invocations, the full validation prompt, the git commit/push bash block, and a complete validation-log.md markdown template, covering the common cases with real tool calls and commands rather than vague direction.	5 / 5
Workflow Clarity	The process is explicitly sequenced — the 7-step completion protocol plus Phases 1–3 with sub-steps 3a–3f — with concrete checkpoints ('Wait for all agents to return', idempotency guards, sentinel-must-be-final-commit) and an error-recovery path for crash-mid-write, and the destructive edits are gated by explicit 3-subagent consensus + expert review + classification, so the destructive-operation cap does not apply.	5 / 5
Progressive Disclosure	No bundle files exist (references/, scripts/, assets/ are absent) and all content lives inline in a ~250-line SKILL.md with good section headers but no one-level-deep references; content that could be split (the full classification rubric, the log template) is inlined, fitting the 'some structure but content that should be separate is inline' anchor.	3 / 5
	Total	16 / 20 Passed

Description

55%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 concrete and highly distinct in its niche, but it is written for a headless dispatcher rather than a user: it lacks any 'Use when...' trigger guidance and relies on internal jargon instead of natural trigger terms. Specificity and distinctiveness are strong; completeness and trigger-term quality are the weak points.

Suggestions

Add an explicit trigger clause (e.g. 'Use when the dispatcher has completed spec-planning for <feature> and needs the spec validated before implementation') so the 'when' is stated, not implied.

Pair the jargon with at least one natural phrase a user might say (e.g. 'spec validation', 'check the spec for gaps') to broaden trigger-term coverage beyond system-internal terms.

Consider stating the trigger condition in terms of the sentinel/state-machine context so it reads as guidance rather than purely a mechanism description.

Dimension	Reasoning	Score
Specificity	Quotes concrete actions — 'Validates a mainspec and its slices via 3-subagent consensus plus expert review', 'applies impactful fixes directly to the spec files', 'Touches specs/<feature>/.validated as its final committed action' — listing several specific actions rather than vague language, but it describes a single workflow rather than a broad enumeration of capabilities, so it stops short of the comprehensive-anchor 5.	4 / 5
Completeness	The 'what' is clear (validate spec via consensus + expert review, then apply impactful fixes), but there is no 'Use when...' clause or equivalent explicit trigger guidance — the skill is described as dispatcher-invoked headless with only weakly implied timing, so per the missing-trigger cap it cannot exceed 3.	3 / 5
Trigger Term Quality	The terms present ('mainspec', 'slices', '3-subagent consensus', 'sentinel', '.validated') are system-internal jargon; there are a couple of generic keywords ('validates', 'fixes') but none of the natural phrases a user would actually say, matching the 'one or two generic keywords; missing the natural phrases' anchor rather than the no-keywords anchor 1.	2 / 5
Distinctiveness Conflict Risk	The mainspec/slices + 3-subagent-consensus + sentinel framing carves a clear niche specific to this spec-validation pipeline with distinct triggers and minimal overlap risk with other skills, fitting the 'clear niche with distinct triggers' anchor.	5 / 5
	Total	14 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: tdg-ninja/context-specs-factory-ai
Path: skills/sdd/spec-validate/SKILL.md
Commit: 3a7a725

Reviewed: about 20 hours ago

Table of Contents

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.