clarity-gate

Pre-ingestion verification for epistemic quality in RAG systems. Ensures documents are properly qualified before entering knowledge bases. Produces CGD (Clarity-Gated Documents) and validates SOT (Source of Truth) files.

Quality

46%

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/clarity-gate/SKILL.md

Quality

Content

46%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

This skill demonstrates strong domain expertise and provides genuinely actionable verification guidance with concrete examples and clear before/after patterns. However, it suffers significantly from verbosity—inlining format specification details, extensive validation code tables, and a lengthy changelog that should live in referenced files. The workflow for actually performing a Clarity Gate review is described conceptually but lacks a clear, sequenced step-by-step process with explicit validation checkpoints.

Suggestions

Move validation code tables (E-TB*, W-HC*, E-ST*, etc.), claim ID format details, and document hash algorithm specifics into the referenced CLARITY_GATE_FORMAT_SPEC.md, keeping only a brief summary in SKILL.md with a clear pointer to the spec.

Move the changelog to a separate CHANGELOG.md file and remove it from SKILL.md entirely.

Add a clear, numbered top-level workflow section (e.g., '## How to Run a Clarity Gate Review') with explicit steps: 1) Scan document with 9 points, 2) Generate findings report, 3) Run Round A confirmation, 4) Run Round B HITL verification, 5) Produce CGD output, 6) Validate with document_hash.py, with validation checkpoints between steps.

Reduce the CGD Output Format section to a single concise template with a reference to the FORMAT_SPEC for field-level details, rather than re-documenting all field semantics inline.

Dimension	Reasoning	Score
Conciseness	The skill is extremely verbose at ~500+ lines. It extensively documents validation codes, changelog history, claim ID collision probabilities, and format specification details that could be in referenced files. Much of this is reference material that should live in the FORMAT_SPEC.md it already references, not in the main SKILL.md. The changelog alone is ~40 lines of content that adds no actionable guidance.	2 / 5
Actionability	The skill provides concrete examples throughout: executable script commands with expected outputs, clear before/after tables for each verification point, a complete CGD output template, and specific fix patterns. Minor gaps exist—the overall workflow for actually running a full Clarity Gate review end-to-end could be more explicitly stepped out with exact commands rather than described conceptually.	4 / 5
Workflow Clarity	The verification hierarchy diagram and Round A/B process provide a rough sequence, but the overall workflow lacks explicit step-by-step sequencing with validation checkpoints. There's no clear 'Step 1, Step 2, Step 3' for performing a complete Clarity Gate review. The 9 verification points are listed but not sequenced into a workflow with validation gates between them. For a process involving document manipulation and integrity hashing, the absence of explicit validation-then-proceed steps is notable.	3 / 5
Progressive Disclosure	The skill references CLARITY_GATE_FORMAT_SPEC.md but then proceeds to inline massive amounts of the spec's content (validation codes, claim ID formats, exclusion block rules, SOT table validation). No bundle files are provided, so we can't verify the referenced spec exists, but regardless, the SKILL.md duplicates what should be in that spec file. The changelog, related projects, and detailed format specification content should be in separate files, leaving the SKILL.md as a concise overview.	2 / 5
	Total	11 / 20 Passed

Description

46%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 targets a narrow, specialized domain (RAG pre-ingestion verification) which gives it reasonable distinctiveness, but it relies heavily on jargon that users are unlikely to use naturally. It lacks an explicit 'Use when...' clause and the core actions beyond producing CGD/SOT files remain vague, making it difficult for Claude to confidently select this skill from a large pool.

Suggestions

Add an explicit 'Use when...' clause with natural trigger phrases, e.g., 'Use when adding documents to a knowledge base, checking document quality before ingestion, or preparing files for a RAG pipeline.'

Replace or supplement jargon with natural language equivalents—e.g., explain what CGD and SOT mean in plain terms and include user-facing phrases like 'document quality check', 'knowledge base preparation', or 'source verification'.

Specify the concrete verification actions performed, such as 'checks factual consistency, identifies ambiguous claims, flags unsupported assertions, and scores document clarity'.

Dimension	Reasoning	Score
Specificity	Names the domain (RAG pre-ingestion verification) and a couple of concrete outputs (CGD documents, SOT file validation), but the core actions like 'ensures documents are properly qualified' remain vague—what does qualification entail specifically?	3 / 5
Completeness	The 'what' is partially described (verification, producing CGD, validating SOT), but there is no explicit 'when' clause. There's no 'Use when...' guidance telling Claude when to select this skill, which caps this dimension at 3 per the rubric guidelines.	3 / 5
Trigger Term Quality	Uses highly specialized jargon like 'epistemic quality', 'CGD', 'SOT', and 'pre-ingestion verification' that users are unlikely to naturally say. Missing common terms users might use like 'check documents', 'knowledge base quality', 'document review', or 'RAG pipeline'.	2 / 5
Distinctiveness Conflict Risk	The niche is fairly distinct—pre-ingestion document verification for RAG systems with specific output formats (CGD, SOT). The specialized terminology and narrow focus reduce overlap risk, though it could potentially conflict with general document validation or RAG-related skills.	4 / 5
	Total	12 / 20 Passed

Validation

81%

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 — 9 / 11 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (714 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	9 / 11 Passed

Repository: administrakt0r/AI-Agents-Safe-Coding-Skills
Path: skills/clarity-gate/SKILL.md
Commit: 57c135b

Reviewed: 2 days 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.