Write technical docs (architecture overviews, feature designs, runbooks, getting-started guides, READMEs, tech-debt notes, how-tos, implementation plans, RFCs, and more). Asks the doc type and audience first, runs targeted clarifying questions one at a time, then drafts in plain language with structured markdown and pastel mermaid diagrams where they help. Output is humanised, with no AI tells.
Personal Claude Code plugin marketplace.
Add the marketplace:
/plugin marketplace add LorcanChinnock/lorcan-claude-marketplace
Then browse and install plugins:
/plugin
Or install a specific plugin directly:
/plugin install <plugin-name>@lorcan-claude-marketplace
/plugin marketplace update lorcan-claude-marketplace
Clone and point Claude at the working copy instead of GitHub:
git clone https://github.com/LorcanChinnock/lorcan-claude-marketplace.git
/plugin marketplace add /absolute/path/to/lorcan-claude-marketplace
describe-pr — write a conventional-commits title and PR description for the current branch from the raw diff vs base. Never pushes, commits, or opens a PR.humanize — rewrite text and code to remove signs of AI-generated writing. Prose skill (humanize-text) handles inflated significance, promotional language, em-dash overuse, AI vocabulary, bolded-header bullets, sycophantic openers. Code skill (humanize-code) renames verbose identifiers, simplifies comments and docstrings, and cleans up log / error / user-facing strings for non-native English readers.handle-review — structured workflow for responding to code review or other critical feedback. Enforces verify-before-implement, reasoned push-back, one-item-at-a-time execution, and no performative agreement.tech-docs — write technical docs (architecture overviews, feature designs, runbooks, getting-started guides, READMEs, tech-debt notes, how-tos, implementation plans, RFCs, and more). Asks the doc type and audience first, runs targeted clarifying questions one at a time, then drafts in plain language with structured markdown and pastel mermaid diagrams where they help. Output is humanised.Register each new plugin by adding an entry to the plugins array in .claude-plugin/marketplace.json.
version lives in its own .claude-plugin/plugin.json (not in the marketplace entry). The Claude Code docs note that for relative-path marketplaces the spec prefers the marketplace entry, but plugin.json wins silently when both are set and keeping it with the plugin keeps the bump local to the change. Only ever set the version in one place.version in plugin.json whenever you change that plugin's skills, agents, hooks, commands, or other user-facing behavior. Patch for small tweaks, minor for new features, major for breaking changes.claude plugin validate . from the repo root before pushing. The marketplace and every plugin should validate cleanly.MIT.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
npx claudepluginhub lorcanchinnock/lorcan-claude-marketplaceWriting quality skills: humanize-text removes AI tells from prose, humanize-code cleans up verbose identifiers and padded comments, and write-doc authors technical documentation (AOs, runbooks, READMEs, RFCs, and more) with audience-aware plain language and pastel Mermaid diagrams.
Local-only git workflow skills for managing stacked branches. Includes tools for splitting large branches into independently mergeable stacked PRs, and for rebasing entire stacks cleanly using git's --update-refs flag.
PR lifecycle skills: describe-pr writes a conventional-commits title and humanised PR body from the raw diff vs base. handle-review processes incoming code review feedback — verify-before-implement, reasoned push-back, one item at a time, no performative agreement. Neither skill pushes, merges, or posts to GitHub.
Use when responding to code review or other critical feedback (PR comments, design review, written critique). Enforces verify-before-implement, reasoned push-back, one-item-at-a-time execution, and no performative agreement. Includes read-only gh helpers for PR context.
Creative design skills for ideation and direction. The creative-director skill acts as a senior UX/creative director: it live-researches current AI design tropes to avoid and fresh references, then produces opinionated, conceptually rich design direction.
Make your AI agent code with your project's architecture, rules, and decisions.
Access official Microsoft documentation, API references, and code samples for Azure, .NET, Windows, and more.
Upstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.
Build and maintain an LLM-curated personal knowledge base in your project — Andrej Karpathy's LLM Wiki pattern, designed to scale to thousands of pages without becoming a context bottleneck. Now with an optional compiled graph layer for typed, provenance-backed relationships.
Connect to Atlassian products including Jira and Confluence. Search and create issues, access documentation, manage sprints, and integrate your development workflow with Atlassian's collaboration tools.
AI-powered wiki generator for code repositories. Generates comprehensive, Mermaid-rich documentation with dark-mode VitePress sites, onboarding guides, deep research, and source citations. Inspired by OpenDeepWiki and deepwiki-open.