axiom-audit-memory

Use when the user mentions memory leak prevention, code review for memory issues, or proactive leak checking.

Quality

63%

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 ./axiom-codex/skills/axiom-audit-memory/SKILL.md

Quality

Content

77%

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 with a clear, validated multi-phase workflow and concrete grep/CLI commands. Its main weaknesses are a monolithic single-file structure with no progressive disclosure into bundle files, and modest verbosity in impact/notes commentary.

Suggestions

Split the six detection patterns and/or the output template into a reference file (e.g. references/patterns.md) and link to it from SKILL.md so the main file stays a concise overview.

Trim or consolidate repeated per-pattern Impact/Note lines and the restated observer/Combine guidance to improve token efficiency.

Move the Field Crash Correlation table and Related skill pointers into a short references section to reduce inline length while keeping them discoverable.

Dimension	Reasoning	Score
Conciseness	The body is mostly lean, actionable grep patterns and commands, but per-pattern Impact/Note lines and some repetition (e.g. observer/Combine guidance restated across patterns) could be tightened. It avoids explaining basic concepts Claude knows, so it is not a 1, but the length and incidental commentary keep it from a 3.	2 / 3
Actionability	Concrete copy-paste Grep regexes, explicit file-exclusion globs, a runnable `xcsym crash --format=summary` command, and a full output template give fully executable, specific guidance. This matches the 'fully executable code/commands; copy-paste ready' anchor.	3 / 3
Workflow Clarity	Five numbered phases with sub-steps, explicit validation checkpoints ("use Read to verify the surrounding context before reporting", "Require evidence from the Phase 1 map — don't speculate"), and a cross-reference/compound-severity phase provide a clear sequence with feedback loops and checklists, matching the top anchor.	3 / 3
Progressive Disclosure	Content is well-sectioned by phase but lives entirely in one ~250-line monolithic SKILL.md with no bundle files (references/scripts/assets absent) and no one-level-deep references for the six detection patterns, the scoring tables, or the output template. The under-50-lines exception does not apply, so this fits 'some structure but content that should be separate is inline'.	2 / 3
	Total	10 / 12 Passed

Description

50%

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 gives clear, natural trigger guidance but leans entirely on the "when" and leaves the "what" implied, with a narrow set of trigger phrasings. It is competent but could be sharper and more concrete.

Suggestions

Lead with a concrete statement of what the skill does, e.g. "Audits Swift codebases for memory leak patterns (timers, observers, closure captures, delegate cycles, unbounded collections). Use when..."

Broaden trigger terms to include natural variations users say: "memory leak", "retain cycle", "memory growth", "OOM/jetsam", "leak detection".

Trim the generic "code review for memory issues" phrasing or scope it to avoid overlap with general code-review and concurrency skills.

Dimension	Reasoning	Score
Specificity	The phrase "memory leak prevention, code review for memory issues, or proactive leak checking" names the domain and a few activities but does not enumerate concrete actions (e.g. detect missing invalidate calls, find unbounded collections), matching the 'names domain and some actions, but not comprehensive' anchor. It is not a 1 because it is not vague, and not a 3 because no multiple specific concrete actions are listed.	2 / 3
Completeness	The "Use when..." clause explicitly answers when, but what the skill does is only implied by "leak checking" rather than stated (e.g. "audits Swift code for memory leak patterns"). One half is explicit and the other implied, matching the score-2 anchor; it is not 3 because both what and when are not clearly answered.	2 / 3
Trigger Term Quality	Terms like "memory leak prevention", "memory issues", and "leak checking" are natural, but common variations a user might actually say (e.g. "retain cycle", "memory growth", "OOM", "leak detection") are missing, fitting 'some relevant keywords but missing common variations'. Not a 3 because coverage of natural terms is narrow.	2 / 3
Distinctiveness Conflict Risk	Memory leak auditing is a recognizable niche, but "code review for memory issues" overlaps with general code-review and concurrency skills (the body itself cross-references axiom-concurrency), fitting 'somewhat specific but could still overlap with similar skills'. Not a 3 because conflict with adjacent auditors is plausible.	2 / 3
	Total	8 / 12 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: CharlesWiltgen/Axiom
Path: axiom-codex/skills/axiom-audit-memory/SKILL.md
Commit: ea3be7c

Reviewed: 1 day 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.