CtrlK
BlogDocsLog inGet started
Tessl Logo

akf-trust-metadata

The AI native file format. EXIF for AI — stamps every file with trust scores, source provenance, and compliance metadata. Embeds into 20+ formats (DOCX, PDF, images, code). EU AI Act, SOX, HIPAA auditing.

52

Quality

59%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/akf-trust-metadata/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

68%Scale 1-5

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

A reasonably well-structured skill that provides concrete CLI commands for AI file metadata stamping and auditing. Its main weakness is the lack of an explicit end-to-end workflow with validation checkpoints (e.g., stamp → verify → audit), which is important for metadata operations that affect compliance. The skill also undersells its claimed breadth (20+ formats, detailed compliance) without pointing to deeper documentation.

Suggestions

Add an explicit numbered workflow with a validation step: e.g., 1. Stamp the file, 2. Read/inspect to verify metadata was applied, 3. Audit for compliance — with example output showing expected results.

Include sample command output for at least `akf read` or `akf inspect` so Claude knows what to expect and can verify success.

Reference or link to format-specific guidance for the 20+ supported formats (DOCX, PDF, images, code) — even a brief table of format → embedding notes would improve progressive disclosure.

DimensionReasoningScore

Conciseness

Generally lean and efficient. The opening analogy ('Every photo has EXIF. Every song has ID3.') is a minor indulgence but adds context quickly. The evidence examples list is slightly verbose but still useful. Overall respects Claude's intelligence.

4 / 5

Actionability

Provides concrete, copy-paste CLI commands for stamping, reading, inspecting, and auditing. However, it lacks a complete workflow example showing end-to-end usage (e.g., stamp then verify), and there's no example output shown for any command, which leaves minor gaps.

4 / 5

Workflow Clarity

The sections imply a sequence (create/modify → stamp, before modifying → read, then audit), but there's no explicit numbered workflow tying these together. For a tool that modifies file metadata, there's no validation step to confirm the stamp was applied correctly (e.g., stamp then read to verify). Missing verification caps this at 3.

3 / 5

Progressive Disclosure

For a relatively short skill with no bundle files, the content is well-organized into clear sections (when to use, after creating, before modifying, compliance, classification, install, links). External links are provided for deeper exploration. Minor gap: no references to separate docs for advanced features like format-specific embedding details for 20+ formats mentioned in the description.

4 / 5

Total

15

/

20

Passed

Description

50%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 has a clear domain focus on AI provenance and compliance metadata embedding, with specific format and regulation references that help distinguish it. However, it reads like marketing copy ('The AI native file format') rather than a functional skill description, and critically lacks any 'Use when...' guidance for Claude to know when to select it. The trigger terms lean heavily toward jargon rather than natural user language.

Suggestions

Add an explicit 'Use when...' clause with natural trigger phrases, e.g., 'Use when the user asks about file provenance, AI content tracking, compliance metadata, audit trails, or embedding trust information into documents.'

Replace the marketing-style opening ('The AI native file format. EXIF for AI') with a concrete action-oriented statement, e.g., 'Embeds AI provenance metadata, trust scores, and compliance audit trails into files.'

Include more natural user-facing synonyms such as 'track file origin', 'content authenticity', 'audit trail', 'metadata stamping', or 'regulatory compliance documentation'.

DimensionReasoningScore

Specificity

Lists several specific actions: stamps files with trust scores, source provenance, compliance metadata, embeds into 20+ formats. Mentions specific formats (DOCX, PDF, images, code) and specific compliance frameworks (EU AI Act, SOX, HIPAA). Minor gaps in explaining exactly what operations are performed beyond embedding.

4 / 5

Completeness

Has a somewhat clear 'what' (embeds trust scores, provenance, and compliance metadata into files), but there is no 'when' clause at all. The description reads more like marketing copy than a functional skill description. Per rubric guidelines, missing 'Use when...' clause caps completeness at 3, and the vague/marketing-style 'what' pushes it to 2.

2 / 5

Trigger Term Quality

Includes some relevant keywords like 'EXIF', 'trust scores', 'provenance', 'compliance', 'DOCX', 'PDF', 'EU AI Act', 'SOX', 'HIPAA', and 'auditing'. However, these are more technical/jargon terms than natural phrases a user would say. Missing natural trigger phrases like 'metadata', 'track origin', 'audit trail', or 'file certification'.

3 / 5

Distinctiveness Conflict Risk

The combination of AI provenance metadata, trust scores, and compliance framework references (EU AI Act, SOX, HIPAA) creates a fairly distinct niche. Minor overlap risk with general PDF/document processing skills or generic compliance skills, but the specific focus on AI provenance metadata is distinctive.

4 / 5

Total

13

/

20

Passed

Validation

90%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation10 / 11 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

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

Warning

Total

10

/

11

Passed

Repository
administrakt0r/AI-Agents-Safe-Coding-Skills
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.