Skill Claude Code
Creates or updates the "Voice & Messaging" section in a client''s CLAUDE.md file, syncing voice patterns, vocabulary.
not rated 36
+1 1mo ago A 136 tokens
original MIT
36Stars on the repository
169Mods indexed here, across every type
1mo agoLast push, which is what freshness is scored on
MITLicence, which decides whether bodies are shown
Skill Claude Code
Triage decision tree for design incidents flagged by /design-reviewer.
not rated 36
+1 1mo ago A SkillSpector: pass 173 tokens
original MIT
Skill Claude Code
Reviews implemented React/Tailwind/CSS output for design quality across 5 dimensions scored 0-4: anti-patterns.
not rated 36
+1 1mo ago A SkillSpector: pass 158 tokens
original MIT
eval-harness
04Skill Claude Code
Deterministic programmatic gate for Genesys skill outputs and client deliverables. Stdlib Python evaluator that scores markdown artifacts against weighted rubric.json files — section presence, sentence/word counts, regex patterns, keyword anti-pattern guards, frontmatter validation, citation tag presence. No LLM…
not rated 36
+1 1mo ago A SkillSpector: pass 144 tokens
original MIT
Skill Claude Code needs its repo
Reviews any document — strategy doc, spec, brief, launch plan, proposal — through a product POV across 6 dimensions: target-problem clarity, persona focus, metric quality (anti-vanity), differentiation defensibility, scope discipline, and shipping-as-experiment framing. Produces a scored product-review-report with…
not rated 36
+1 1mo ago A SkillSpector: pass 155 tokens
original MIT
Skill Claude Code needs its repo
Reviews client proposals, statements of work, and engagement scope documents for scope creep, mission drift, and unrealistic delivery commitments. Catches the common B2B SaaS consulting failure modes — open-ended deliverable lists, unbounded revision counts, undefined acceptance criteria, dependencies on undelivered…
not rated 36
+1 1mo ago B 150 tokens
original MIT
Skill Claude Code
Maintains the central registry of all skills with dependencies, trigger phrases, review gates, and chaining.
not rated 36
+1 1mo ago A SkillSpector: warn 96 tokens
original MIT
Skill Claude Code
Reviews SKILL.md files for structural quality across 8 dimensions (frontmatter, triggers, process, inputs, outputs.
not rated 36
+1 1mo ago A SkillSpector: pass 111 tokens
original MIT
Skill Claude Code
Reviews content output for voice quality and brand compliance across 8 dimensions (D8 is LinkedIn-only) including the 100 Posts Test.
not rated 36
+1 1mo ago A 124 tokens
original MIT
connect-mcp
10Skill Claude Code
Step-by-step runbook to connect any custom MCP to Claude — choosing the source (official plugin / community / custom HTTP / local stdio), deciding the mounting tier, wiring credentials, registering the server, introspecting its tools, and integrating it into skills. Operationalizes the mcp-on-demand.md policy with the…
not rated 36
+1 1mo ago C 129 tokens
original MIT
Skill Claude Code
Creates branded Google Docs, Slides, and Sheets via GDrive MCP scripts, automatically routing documents to the.
not rated 36
+1 1mo ago A SkillSpector: pass 120 tokens
original MIT
Skill Claude Code
Creates Mermaid diagrams for flowcharts, architecture maps, sequence diagrams, and process visualizations. Produces.
not rated 36
+1 1mo ago A SkillSpector: warn 110 tokens
original MIT
Skill Claude Code
Scaffolds a Claude Code / Cowork plugin bundle from a short YAML spec or from an existing agent definition''s.
not rated 36
+1 1mo ago A SkillSpector: warn 159 tokens
original MIT
Skill Claude CodeCodex
Run a web task as a multi-step browser loop on MCPs you already pay for. No new runtime, no API double-billing, no new always-on MCP. Two v1 modes: monitor (watch a page, report a grounded diff) and QA (walk a site, check CTAs/links/schema, flag mismatches).
not rated 36
+1 1mo ago A 206 tokens
original MIT
discover
15Skill Claude Code
Proactive skill and automation discovery. Scans 6 data sources (sessions, Linear, Slack, Gmail, Granola, Calendar).
not rated 36
+1 1mo ago A 100 tokens
original MIT
experiment
16Skill Claude Code
Tracks structured experiments for content, messaging, and workflow iterations in a SQLite database. Logs hypotheses.
not rated 36
+1 1mo ago A 121 tokens
original MIT
learn
17Skill Claude Code
Process raw source material (transcripts, docs, articles, competitor pages, meeting notes) into structured context.
not rated 36
+1 1mo ago A SkillSpector: warn 74 tokens
original MIT
level
18Skill Claude Code
Assesses a Claude Code environment''s maturity on a fun 0-10 ladder calibrated to the 4-systems.
not rated 36
+1 1mo ago A SkillSpector: pass 169 tokens
original MIT
Skill Claude Code
Scans skill-usage.jsonl + recall.db transcripts for opportunities to create new skills, deprecate unused ones, or formalize manual patterns into workflows. Detects three signal classes: repeated manual sequences not covered by existing skills, skills used <2× in 30 days (deprecation candidates), high-friction sessions…
not rated 36
+1 1mo ago A SkillSpector: warn 136 tokens
original MIT
Skill Claude Code
Dynamically guides a person through cloning + setting up the claude-code-marketing-quickstart.
not rated 36
+1 1mo ago A SkillSpector: warn 164 tokens
original MIT
runbook
21Skill Claude Code
Synthesizes documentation sources (SKILL.md files, MCP configs, code comments, client docs) into structured.
not rated 36
+1 1mo ago A SkillSpector: warn 119 tokens
original MIT
steal
22Skill Claude Code
Systematically extracts and adapts reusable skills, frameworks, benchmarks, and patterns from external GitHub.
not rated 36
+1 1mo ago A SkillSpector: pass 148 tokens
original MIT
wiki
23Skill Claude Code
Per-client-folder hygiene and semantic drift detection. Runs structured checks inside a single client engagement.
not rated 36
+1 1mo ago A SkillSpector: pass 129 tokens
original MIT
batch-run
24Skill Claude Code
Fan out any skill across N entities in parallel. Takes a skill name + entity list (products, competitors, pages.
not rated 36
+1 1mo ago A SkillSpector: pass 136 tokens
original MIT
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: