CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

codebase-audit-pre-push

Deep audit before GitHub push: removes junk files, dead code, security holes, and optimization issues. Checks every file line-by-line for production readiness.

Quality

67%

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 ./plugins/AI-Agents-Safe-Coding-Skills-claude/skills/codebase-audit-pre-push/SKILL.md

The canonical home for this skill is codebase-audit-pre-push in administrakt0r/AI-Agents-Safe-Coding-Skills

SKILL.md
Quality
Evals
Security

Quality

Content

48%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.

A well-sequenced, thorough audit procedure with clear sections and a final validation step, but it is token-heavy: it re-teaches engineering fundamentals Claude already knows, repeats itself, and inlines catalogs and templates that belong in reference files. Actionable direction is present but lacks executable search commands and a practical method for auditing large codebases.

Suggestions

Trim the skill to non-obvious audit criteria and move the pattern catalogs (junk-file list, security/scalability/architecture checklists) and the report template into references/ files (e.g., references/file-patterns.md, references/security-checklist.md), keeping SKILL.md as a concise overview with one-level-deep links.

Remove the redundant 'Key Principles' section that repeats 'Your Job' verbatim, and cut the 35-line sample report to a short skeleton showing only the section headings.

Add executable discovery commands (e.g., grep patterns for secrets and debug statements, git status/git clean --dry-run for untracked junk, a dry-run flag before deletions) so 'review the entire codebase file by file' becomes a practical, repeatable procedure — including how to handle deleting node_modules/ without breaking the step-10 'run the app' verification.

DimensionReasoningScore

Conciseness

Roughly 230 lines largely enumerate standard engineering knowledge Claude already has (use ===, bcrypt over MD5, N+1 queries, what dead code is), and "Key Principles" repeats instructions verbatim from "Your Job" ("Read the code thoroughly" / "Fix issues immediately"), plus a 35-line sample report. Not 1: it is organized checklist content, not discursive padding; not 3: the sheer volume of re-taught basics and duplicated sections is noticeably wasteful.

2 / 5

Actionability

Concrete specifics exist (exact junk patterns like .DS_Store and npm-debug.log*, "Run npm audit", a full report template), but there are no executable commands for finding issues (grep/git patterns) and "Review the entire codebase file by file" gives no practical method. Not 4: the guidance is checklists of what to look for, not executable steps; not 2: the pattern lists and report format are genuinely concrete guidance.

3 / 5

Workflow Clarity

Clear numbered 1-10 sequence with a final verification step ("run the app... tests pass... no regressions") and a destructive-op checkpoint ("If uncertain about removing something, ask the user"), satisfying the destructive-operations validation rule. Not 5: validation is only end-of-process, not per-step, and there is a gap where deleting node_modules/ in step 1 would break "run the app" in step 10 without reinstall guidance.

4 / 5

Progressive Disclosure

Well-organized section headers and no nested references, but the entire audit is inlined in one monolithic file with no references/ directory — the junk-file pattern list, security checklist, and output template are natural candidates for separate reference files. Not 4: content that clearly belongs in separate files (pattern catalogs, report format) is fully inline; not 2: structure is good, not minimal.

3 / 5

Total

12

/

20

Passed

Description

71%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 solid description with concrete action categories, natural trigger terms, and an explicit when-context, written in third person. It is somewhat generic relative to the crowded code-review space and the when-clause covers only a single trigger.

DimensionReasoningScore

Specificity

Lists several concrete actions ("removes junk files, dead code, security holes, and optimization issues") but "optimization issues" and "production readiness" are generic, leaving minor gaps in coverage. Not 5: the action list is not comprehensively concrete; not 3: it clearly enumerates multiple specific actions rather than just 1-2.

4 / 5

Completeness

Both what (remove junk, dead code, security holes, optimization issues) and when ("before GitHub push") are present, so the missing-trigger cap at 3 does not apply. Not 5: the when is a single narrow trigger, less explicit than a full "Use when..." clause covering phrases like making a repo public or pre-deployment review.

4 / 5

Trigger Term Quality

Includes natural phrases users would say ("audit", "GitHub push", "security holes", "dead code", "junk files"). Not 5: missing common variations like "code review", "clean up the code", "pre-deploy", "lint".

4 / 5

Distinctiveness Conflict Risk

"Deep audit" and "production readiness" overlap heavily with generic code-review and security-review skills; "before GitHub push" is the only real differentiator. Could still trigger for the wrong skill when a user asks for a general code review.

3 / 5

Total

15

/

20

Passed

Validation

93%

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

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

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

Warning

Total

15

/

16

Passed

Repository

administrakt0r/AI-Agents-Safe-Coding-Skills

Path

plugins/AI-Agents-Safe-Coding-Skills-claude/skills/codebase-audit-pre-push/SKILL.md

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.