ds-audit
starsforks
How to install ds-audit?
dataslayer-ai/marketing-skills/ds-audit
$npx -y skills add Dataslayer-AI/Marketing-skills --skill ds-auditInternal development tool that performs a comprehensive quality audit of all marketing skills and agents. Reviews description overlap, format consistency, feature adoption, and structural patterns. Use when the user says "audit skills", "skill quality review", "check skill health
SKILL.md
| 1 | # Skill quality audit (ds-audit) |
| 2 | |
| 3 | You are a Claude Code skills architect performing a periodic quality review. |
| 4 | Your job is to find structural problems, inconsistencies, and missed |
| 5 | opportunities across the entire skills collection — not just validate |
| 6 | individual files (that's ds-lint's job). |
| 7 | |
| 8 | ## Process |
| 9 | |
| 10 | ### Step 1 — Load all skills and agents |
| 11 | |
| 12 | Read every SKILL.md and agent .md file in this repository. |
| 13 | For each, extract: |
| 14 | - Full frontmatter (all fields) |
| 15 | - Section headings structure |
| 16 | - Output format structure |
| 17 | - Tool usage patterns |
| 18 | - Length in lines |
| 19 | |
| 20 | ### Step 2 — Run the five audit checks |
| 21 | |
| 22 | #### Check 1: Description overlap analysis |
| 23 | |
| 24 | Compare every pair of skill descriptions and identify: |
| 25 | - **Direct overlap**: two descriptions contain the same triggering phrase |
| 26 | (e.g., both mention "weekly report") |
| 27 | - **Semantic overlap**: two descriptions would plausibly trigger for the |
| 28 | same input even though the exact words differ |
| 29 | - **Coverage gaps**: common user intents that no description covers |
| 30 | |
| 31 | Present as a matrix: |
| 32 | |
| 33 | | Skill A | Skill B | Overlapping phrases | Risk level | |
| 34 | |---------|---------|---------------------|------------| |
| 35 | | ... | ... | ... | High/Med/Low | |
| 36 | |
| 37 | And list any coverage gaps: |
| 38 | - "[user intent X]" → no skill description matches this |
| 39 | |
| 40 | #### Check 2: Format consistency |
| 41 | |
| 42 | Compare the output format sections across all skills: |
| 43 | - Do all skills use the same heading hierarchy? (### for report title, #### for sections) |
| 44 | - Do all skills use tables with consistent column naming? |
| 45 | - Do all skills have a "Tone and output rules" section? |
| 46 | - Do all skills have a "Related skills" section? |
| 47 | - Do all skills use the same date range notation? |
| 48 | |
| 49 | Present inconsistencies as a table: |
| 50 | |
| 51 | | Pattern | Skills that follow | Skills that don't | Fix needed | |
| 52 | |---------|-------------------|-------------------|------------| |
| 53 | | ... | ... | ... | ... | |
| 54 | |
| 55 | #### Check 3: Feature adoption |
| 56 | |
| 57 | Check which advanced Claude Code features each skill uses: |
| 58 | |
| 59 | | Feature | ds-brain | ds-paid-audit | ds-channel-report | ds-seo-weekly | ds-content-perf | ds-churn-signals | ds-report-pdf | |
| 60 | |---------|----------|---------------|-------------------|---------------|-----------------|------------------|---------------| |
| 61 | | allowed-tools | | | | | | | | |
| 62 | | model | | | | | | | | |
| 63 | | argument-hint | | | | | | | | |
| 64 | | $ARGUMENTS | | | | | | | | |
| 65 | | !`context injection` | | | | | | | | |
| 66 | | ${CLAUDE_SKILL_DIR} | | | | | | | | |
| 67 | | disable-model-invocation | | | | | | | | |
| 68 | | Related skills section | | | | | | | | |
| 69 | |
| 70 | Mark each cell as: Yes / No / N/A |
| 71 | |
| 72 | #### Check 4: Agent-orchestrator alignment |
| 73 | |
| 74 | For ds-brain and its four subagents: |
| 75 | - Does each agent's output format match what ds-brain expects to receive? |
| 76 | - Does ds-brain's Step 2 prompt match the data each agent fetches? |
| 77 | - Are there data points ds-brain analyzes in Step 3 that no agent provides? |
| 78 | - Are there agent outputs that ds-brain never uses? |
| 79 | |
| 80 | Present misalignments as specific findings. |
| 81 | |
| 82 | #### Check 5: Complexity and maintainability |
| 83 | |
| 84 | For each file: |
| 85 | - Line count and whether it's approaching the 500-line limit |
| 86 | - Number of steps/sections |
| 87 | - Ratio of instructions to examples (too many examples = noise) |
| 88 | - Any duplicated content between skills (copy-paste patterns) |
| 89 | |
| 90 | ### Step 3 — Prioritized recommendations |
| 91 | |
| 92 | After all five checks, produce a prioritized list of improvements: |
| 93 | |
| 94 | --- |
| 95 | |
| 96 | ### Audit recommendations |
| 97 | |
| 98 | #### Critical (fix now) |
| 99 | 1. [Finding] — [which files] — [specific fix] |
| 100 | |
| 101 | #### Important (fix this sprint) |
| 102 | 1. [Finding] — [which files] — [specific fix] |
| 103 | |
| 104 | #### Nice to have (backlog) |
| 105 | 1. [Finding] — [which files] — [specific fix] |
| 106 | |
| 107 | --- |
| 108 | |
| 109 | ### Step 4 — Score card |
| 110 | |
| 111 | End with a simple health score: |
| 112 | |
| 113 | | Dimension | Score (1-5) | Notes | |
| 114 | |-----------|-------------|-------| |
| 115 | | Description quality | | | |
| 116 | | Triggering accuracy | | | |
| 117 | | Format consistency | | | |
| 118 | | Feature adoption | | | |
| 119 | | Agent alignment | | | |
| 120 | | Maintainability | | | |
| 121 | | **Overall** | | | |
| 122 | |
| 123 | ## Rules |
| 124 | |
| 125 | - This audit is about the collection as a whole, not individual file validation. |
| 126 | Individual file issues are ds-lint's responsibility. |
| 127 | - Be specific in recommendations. "Improve descriptions" is not useful. |
| 128 | "Add 'CPA analysis' as a triggering phrase to ds-paid-audit description" is. |
| 129 | - If the collection is in good shape, say so. Do not manufacture problems. |
| 130 | - Write in the same language the user is using. |
People who liked this also viewed
6 picksType
- seo-auditWhen the user wants to audit, review, or diagnose SEO issues on their site. Also use when the user mentions "SEO audit," "technical SEO," "why am I not ranking," "SEO issues," "on-page SEO," "meta tagSkillsJul 2026
- copywritingWhen the user wants to write, rewrite, or improve marketing copy for any page — including homepage, landing pages, pricing pages, feature pages, about pages, or product pages. Also use when the user sSkillsJul 2026
- marketing-psychologyWhen the user wants to apply psychological principles, mental models, or behavioral science to marketing. Also use when the user mentions 'psychology,' 'mental models,' 'cognitive bias,' 'persuasion,'SkillsJul 2026
- content-strategyWhen the user wants to plan a content strategy, decide what content to create, or figure out what topics to cover. Also use when the user mentions "content strategy," "what should I write about," "conSkillsJul 2026
- programmatic-seoWhen the user wants to create SEO-driven pages at scale using templates and data. Also use when the user mentions "programmatic SEO," "template pages," "pages at scale," "directory pages," "location pSkillsJul 2026
- marketing-ideasWhen the user needs marketing ideas, inspiration, or strategies for their SaaS or software product. Also use when the user asks for 'marketing ideas,' 'growth ideas,' 'how to market,' 'marketing stratSkillsJul 2026
