CtrlK
BlogDocsLog inGet started
Tessl Logo

axiom-audit-foundation-models

Use when the user mentions Foundation Models review, on-device AI audit, LanguageModelSession issues, or @Generable/guardrail problems — including missing availability checks, manual JSON parsing, and guardrail handling.

68

Quality

82%

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

SKILL.md
Quality
Evals
Security

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.

An exceptionally actionable and well-sequenced audit skill with concrete grep patterns, Swift fixes, and explicit validation checkpoints. Its weaknesses are length/rationale prose that could be trimmed and a monolithic single-file structure with no local progressive-disclosure bundle.

Suggestions

Trim explanatory rationale in the Phase 3 question table and Pattern issue blocks (e.g., 'This is the single largest quality risk...') to leaner statements; the impact is already conveyed by the severity tag.

Move the OS27 error migration table and the per-pattern fix snippets into a local references/ file (e.g. foundation-models-fixes.md) and link one level deep, so SKILL.md stays an overview.

Add the 'Apple Intelligence' trigger synonym to the description and consider an Invocation-Examples trim, since the five example prompts plus the explicit-command line restate the same trigger surface.

DimensionReasoningScore

Conciseness

The ~480-line body is mostly efficient and domain-specific, but it carries substantial explanatory rationale ('This is the single largest quality risk...', 'the model follows the most recent guidance') and rationale prose in the Phase 3/4 tables that could be tightened without losing the actionable signal.

3 / 5

Actionability

Highly executable: runnable grep patterns per pattern, copy-paste Swift fix snippets for every issue, concrete verification steps, and a fully specified output template covering the common audit cases.

5 / 5

Workflow Clarity

A clear five-phase sequence (Map → Detect → Reason → Cross-Reference → Score) with explicit validation checkpoints ('use Read to verify the surrounding context before reporting'), a false-positive checklist, and output-limit rules providing feedback loops for error recovery.

5 / 5

Progressive Disclosure

Good section structure and signaled external references ('See axiom-ai (skills/foundation-models-ref.md) for the full migration table'), but no local bundle files exist in references/scripts/assets and the SKILL.md is a single monolithic file with the bulk of detail inlined rather than split into one-level-deep local reference files.

3 / 5

Total

16

/

20

Passed

Description

87%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 well-crafted description that pairs an explicit 'Use when' trigger clause with concrete, domain-specific detection targets. It is highly distinctive; the only gaps are a slightly generic leading verb and the missing 'Apple Intelligence' synonym.

DimensionReasoningScore

Specificity

Names the domain ('Foundation Models review, on-device AI audit') and several concrete detection targets ('missing availability checks, manual JSON parsing, and guardrail handling'), but the leading verbs 'review/audit' are generic and the list is framed as problem categories rather than a fully comprehensive action set.

4 / 5

Completeness

Explicitly answers both 'what' (detecting Foundation Models issues including the enumerated sub-categories) and 'when' (a 'Use when the user mentions...' clause with concrete trigger phrases).

5 / 5

Trigger Term Quality

Strong, real-world terms a developer would say ('Foundation Models', 'LanguageModelSession', '@Generable', 'guardrail', 'on-device AI'), but it omits the common synonym 'Apple Intelligence' that users often use interchangeably.

4 / 5

Distinctiveness Conflict Risk

Foundation Models / Apple Intelligence on-device AI auditing is a clear, narrow niche with distinct trigger terms and minimal overlap risk against generic skills.

5 / 5

Total

18

/

20

Passed

Validation

100%

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

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
CharlesWiltgen/Axiom
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.