A curated library of Claude Code plugins for data analysis, entity modelling, business operations, brand management, marketing, developer tooling, database design, and DevOps — packaged as a Claude Code marketplace with standalone plugins.
Maintained by @Anthril.
# Add the marketplace
/plugin marketplace add anthril/official-claude-plugins
# Install individual plugins
/plugin install data-analysis@anthril-claude-plugins
/plugin install knowledge-engineering@anthril-claude-plugins
/plugin install business-economics@anthril-claude-plugins
/plugin install package-manager@anthril-claude-plugins
/plugin install plan-completion-audit@anthril-claude-plugins
/plugin install skillops@anthril-claude-plugins
/plugin install brand-manager@anthril-claude-plugins
/plugin install software-development@anthril-claude-plugins
/plugin install ppc-manager@anthril-claude-plugins
/plugin install database-design@anthril-claude-plugins
/plugin install devops@anthril-claude-plugins# Copy one skill to your personal skills directory
cp -r plugins/data-analysis/skills/knowledge-graph-builder ~/.claude/skills/# Load the full marketplace for development
claude --plugin-dir .
# Load a single plugin
claude --plugin-dir ./plugins/data-analysis
# List available skills
/skills
# Run a skill
/knowledge-graph-builder Build a knowledge graph for a consulting firmClaude Code does not auto-refresh marketplaces — it reads from a local cache (~/.claude/plugins/marketplaces/<name>/) that is only re-fetched on demand. If a /plugin view shows you on the latest version when a newer release exists, the cache is stale.
To pick up a new release:
# 1. Re-fetch the marketplace clone (pulls the latest commit)
/plugin marketplace update anthril-claude-plugins
# 2. Update one or more installed plugins
/plugin update software-development@anthril-claude-pluginsThe first command advances the cached marketplace's git HEAD; the second installs the new version into ~/.claude/plugins/cache/anthril-claude-plugins/<plugin>/<version>/ and rewrites the entry in installed_plugins.json.
See CHANGELOG.md for what is in each release before updating.
59 production-ready skills across 11 standalone plugins:
| Skill | Description |
|---|---|
anomaly-detection-rule-builder |
Build rule-based and statistical anomaly detection systems for business metrics — revenue drops, traffic spikes, churn increases, cost overruns |
cohort-analysis-builder |
Design cohort analysis frameworks with SQL queries, visualisation specs, and interpretation guides for retention, revenue, and churn analysis |
data-pipeline-architecture |
Design ETL/ELT pipeline architectures with data flow diagrams, transformation specs, orchestration, and error handling for Supabase and BigQuery |
data-dictionary-generator |
Auto-generate comprehensive data dictionaries from database schemas, CSV files, or API responses with column definitions and Mermaid ERD |
dataset-profiling-quality-audit |
Profile datasets and audit data quality — assess completeness, validity, consistency, uniqueness, timeliness, and accuracy |
| Skill | Description |
|---|---|
business-data-model-designer |
Design complete Supabase/PostgreSQL data models with ERD, SQL migrations, RLS policies, indexes, and triggers |
entity-disambiguation |
Resolve entity ambiguity across data sources — produce canonical records, merge decisions, and sameAs link mappings |
entity-relationship-mapper |
Map business domains to entity-relationship models with Schema.org types, JSON-LD @graph output, and sameAs connections |
knowledge-graph-builder |
Construct knowledge graph specifications for Neo4j, JSON-LD, or Supabase/PostgreSQL JSONB implementation |
| Skill | Description |
|---|---|
unit-economics-calculator |
Calculate CAC, LTV, payback period, contribution margin with scenario analysis for service, SaaS, and hybrid businesses |
market-sizing-tam-estimator |
Estimate TAM, SAM, and SOM using top-down and bottom-up methods with sensitivity analysis, calibrated for Australian markets |
| Skill | Description |
|---|---|
npm-package-audit |
Audit npm packages for publishing quality, cross-OS compatibility, type declarations, build config, security, and CI/CD — produces a scored report with actionable fixes |
cli-ux-audit |
Audit any CLI tool for terminal UX — help text, command structure, error messages, output formatting, discoverability, and accessibility |
| Skill | Description |
|---|---|
plan-completion-audit |
Full-stack audit of a project plan versus actual implementation — verifies plan vs code, types, bugs, security, Supabase schema, RLS, and frontend-backend alignment |
| Skill | Description |
|---|---|
skill-creator |
Create new Claude Code skills with proper frontmatter, directory structure, templates, examples, and supporting files |
skill-evaluator |
Audit an existing skill for quality across eight dimensions — metadata, scope, conciseness, architecture, content, tooling, testing, standards — with a scored markdown report and JSON sidecar |
| Skill | Description |
|---|---|
brand-identity |
Define brand purpose, values, personality, voice, and positioning statement |
brand-guidelines |
Create comprehensive brand guidelines covering typography, colour, imagery, and tone of voice |
target-audience |
Build detailed audience personas with demographics, psychographics, and journey maps |
competitor-analysis |
Analyse competitors across positioning, messaging, visual identity, and market gaps |
logo-brief |
Write a logo design brief with concept direction, usage rules, and file format specs |
color-palette |
Design a brand colour palette with primary, secondary, accent, and semantic colours |
design-tokens |
Generate design tokens for typography, colour, spacing, and elevation in JSON/CSS format |
legal-disclaimers |
Draft legal disclaimers, terms, and compliance notices for Australian businesses |
website-copy |
Write website copy for landing pages, about sections, and CTAs aligned to brand voice |
| Skill | Description |
|---|---|
dead-code-audit |
Detect dead code across 9 languages — JS/TS, Python, Go, Rust, Java, PHP, Ruby, C# — with actionable removal plan |
write-path-mapping |
Map end-to-end write paths from UI to database with framework and database introspection |
| Skill | Description |
|---|---|
oauth-setup |
Walk through OAuth setup for Google and Meta platforms with encrypted vault storage |
gtm-setup |
Set up Google Tag Manager containers, workspaces, and base configuration |
gtm-datalayer |
Design and implement GTM data layer specifications |
gtm-tags |
Create and configure GTM tags, triggers, and variables |
ga4-setup |
Set up GA4 properties, data streams, and base configuration |
ga4-events |
Design and implement GA4 custom events and conversions |
google-ads-account-setup |
Set up Google Ads account structure, billing, and conversion tracking |
google-search-campaign |
Build Google Search campaigns with ad groups, keywords, and ads |
google-pmax-campaign |
Build Google Performance Max campaigns with asset groups and signals |
google-ads-copy |
Write Google Ads copy — headlines, descriptions, and extensions |
display-ad-specs |
Generate display ad specifications and creative briefs |
meta-pixel-setup |
Set up Meta Pixel with base code and standard events |
meta-capi-setup |
Configure Meta Conversions API for server-side tracking |
meta-events-mapping |
Map business events to Meta standard and custom events |
meta-audience-builder |
Build Meta custom and lookalike audiences |
meta-creative-brief |
Write creative briefs for Meta ad campaigns |
meta-ads-copy |
Write Meta ad copy — primary text, headlines, and descriptions |
keyword-research |
Conduct keyword research for PPC campaigns across Google and Meta |
campaign-audit |
Cross-platform campaign audit using all four MCP servers |
utm-builder |
Build UTM parameter conventions and tracking URLs |
landing-page-copy |
Write landing page copy optimised for PPC traffic |
youtube-campaign |
Plan and configure YouTube ad campaigns |
| Skill | Description |
|---|---|
postgres-schema-audit |
Audit any Postgres 13+ schema (Supabase via MCP, or RDS/Cloud SQL/Neon/Railway/self-hosted via a read-only connection) — parallel per-schema sub-agents across ten audit categories, producing evidence-backed findings, an ER diagram, and draft migration SQL |
| Skill | Description |
|---|---|
devops-needs-assessment |
Plain-language DevOps triage for non-experts — scores nine dimensions and names the top three fixes |
cicd-pipeline-audit |
Audit CI/CD pipelines (GitHub Actions, GitLab CI, CircleCI, Azure Pipelines, Jenkins, Bitbucket) — one sub-agent per workflow |
iac-terraform-audit |
Audit Terraform, OpenTofu, Terragrunt, and Pulumi modules — one sub-agent per module |
container-audit |
Audit Dockerfiles and docker-compose files — one sub-agent per Dockerfile |
kubernetes-manifest-audit |
Audit Kubernetes manifests and Helm charts against CIS and NSA/CISA hardening guides |
observability-audit |
Score observability across the four pillars — logs, metrics, traces, alerts/dashboards |
release-readiness-audit |
Pre-production go/no-go gate — migration safety, rollback, monitoring, deploy strategy |
devsecops-supply-chain-audit |
Audit supply chain across every ecosystem detected — pinning, vulnerabilities, secrets, SBOM, signing, branch protection |
sre-reliability-audit |
Assess Site Reliability maturity — SLOs, runbooks, on-call, postmortems, game days |
Every DevOps skill supports three operating modes: static (default), --live (uses gh, kubectl, terraform, cloud CLIs, scanners), and --apply (opt-in remediation with per-change confirmation). Runtime testing (--runtime) is available where applicable with production-name guards.
Every skill in this library includes:
- YAML frontmatter —
name,description(<250 chars),argument-hint,allowed-tools,effort $ARGUMENTS— Accept user input directly (e.g.,/skill-name my business description)ultrathink— Extended thinking enabled for complex analytical skills- Output templates — Structured output format with section headers
- Example outputs — Realistic completed examples with Australian business context
- Utility scripts — Python/Bash helpers for common operations
Select skills also include:
context: fork— Research-heavy skills run in isolated subagent contextpaths— Auto-activation when working with matching file patternsreference.md— Dense reference material (SQL templates, scoring rubrics, lookup tables) extracted to keep SKILL.md under 500 lines- Dynamic context injection — Shell commands that inject project state before the skill runs
- Parallel sub-agents — Independent audit targets (schemas, workflows, modules, Dockerfiles, charts, ecosystems) are audited in parallel for large-repo throughput
Each plugin follows a consistent layout:
plugins/<plugin-name>/
├── .claude-plugin/
│ └── plugin.json # Plugin manifest (name, version, author, keywords)
├── skills/
│ └── <skill-name>/
│ ├── SKILL.md # Main skill instructions (under 500 lines)
│ ├── reference.md # Detailed reference material (where needed)
│ ├── LICENSE.txt # License
│ ├── templates/
│ │ └── output-template.md # Output format skeleton
│ ├── examples/
│ │ └── example-output.md # Realistic completed example
│ └── scripts/
│ └── helper.sh # Utility script (where relevant)
├── hooks/
│ ├── hooks.json # Plugin hooks configuration
│ └── scripts/ # Hook scripts
└── settings.json # Plugin settings
official-claude-plugins/
├── .claude/
│ └── CLAUDE.md # Project instructions for contributors
├── .claude-plugin/
│ └── marketplace.json # Marketplace catalog (11 plugins)
├── plugins/
│ ├── data-analysis/ # Data Analysis & Intelligence (5 skills)
│ ├── knowledge-engineering/ # Knowledge Engineering (4 skills)
│ ├── business-economics/ # Business Economics (2 skills)
│ ├── package-manager/ # Package Manager (2 skills)
│ ├── plan-completion-audit/ # Plan Completion Audit (1 skill)
│ ├── skillops/ # SkillOps (2 skills)
│ ├── brand-manager/ # Brand Manager (9 skills)
│ ├── software-development/ # Software Development (2 skills)
│ ├── ppc-manager/ # PPC Manager (22 skills)
│ ├── database-design/ # Database Design (1 skill)
│ └── devops/ # DevOps (9 skills)
├── scripts/
│ └── check-versions.mjs # Verify marketplace ↔ plugin.json version sync
├── settings.json # Root plugin settings
├── CHANGELOG.md # Version history
├── LICENSE # MIT
└── README.md
Use the built-in skill creator:
/skill-creator customer-churn-predictor — predict churn risk from behavioural signalsOr follow the conventions in .claude/CLAUDE.md to create skills manually.
- SKILL.md has valid YAML frontmatter with
name,description,effort - SKILL.md is under 500 lines
- Uses
$ARGUMENTSfor user input - Description is under 250 characters, front-loaded with key use case
-
effortfield set appropriately (medium,high, ormax) -
pathsfield set if skill should auto-activate on file patterns -
templates/directory has at least one output template -
examples/directory has at least one example output - Dense reference material is in
reference.md, not SKILL.md - Plugin version in
plugin.jsonmatches marketplace entry (runnode scripts/check-versions.mjs) - Tested locally with
claude --plugin-dir .
- Fork the repository
- Create a new skill using
/skill-creator - Place it in the appropriate plugin directory under
plugins/ - Test locally with
claude --plugin-dir . - Submit a pull request
See .claude/CLAUDE.md for detailed development standards.
This project is maintained by Anthril and funded by our sponsors.
MIT