By epologee
Guardrail hooks that push back on common AI reflexes: shifting blame, stopping prematurely, delegating verification, asking for confirmation when none was needed, and em-dashes in prose.
A curated subset of the skills I built while developing applications in Ruby, Swift, Go, JavaScript, Python, Kotlin, and others across various projects. These are the ones I managed to make reusable for colleagues and friends.
I typically prompt in Dutch but write English code, so these skills are a mix of both. The export-skill skill can translate if needed.
claude plugins marketplace add epologee/leclause-skills
claude plugins install <skill-name>@leclause
| Plugin | Command | Auto | Hooks | Description |
|---|---|---|---|---|
| clipboard | /clipboard | Copy the core content of the last answer to the clipboard. /clipboard slack for rich text. | ||
| commit-all-the-things | /commit-all-the-things | Commit all uncommitted changes in the working tree, grouped into logical commits with descriptive messages. | ||
| dont-do-that | ❌ | ✅ | Six guardrail hooks that push back on common AI reflexes: shifting blame, stopping prematurely, delegating verification, and em-dashes in prose. See dont-do-that. | |
| export-skill | /export-skill | Export and sanitize a skill for sharing, stripping PII and checking for security issues. Supports translation and platform porting. | ||
| eye-of-the-beholder | ❌ | ✅ | Catches cramped text, missing margins, and disproportionate spacing in visual layouts (screen, print, responsive). | |
| ground | /ground | ✅ | Verify Claude's recent output with external sources when you challenge accuracy. | |
| gurus | /gurus | Opinionated code review panel with 8 expert perspectives (Beck, Fowler, Uncle Bob, DHH, Metz, Lutke, Hickey, Thoughtbot). | ||
| how-plugins-work | /how-plugins-work | ✅ | Living document explaining how Claude Code plugin naming, skill resolution, and the plugin:skill invocation pattern work. | |
| inspiratie | /inspiratie | ✅ | Online research workflow for unfamiliar topics, design decisions, and evaluating approaches. | |
| rebase-origin-default | /rebase-origin-default | Safely rebase current branch on the remote default branch with staleness check and automatic conflict resolution. | ||
| recap | /recap | Structured status overview of the current session: what we're doing, where we are, what's next. | ||
| recursion | /recursion | Deep research producing atomic improvement plans for the Claude Code workflow. | ||
| rename-suggestion | /rename-suggestion | Suggest a descriptive session name based on conversation context. | ||
| saysay | /saysay | Claude speaks every response aloud. /saysay off to exit. | ||
| screen-recording | /screen-recording | ✅ | Automated screen recordings and demo videos of browser-based features. | |
| self-improvement | /self-improvement | Update CLAUDE.md and skills based on feedback. Detects duplication and extracts large sections into skills. | ||
| testing-philosophy | ❌ | ✅ | Opinionated testing guide covering TDD workflow, Cucumber/Gherkin, flaky test diagnosis, and test suite health. | |
| whywhy | /whywhy [n] | ✅ | Drill N layers deep into a question or goal (default 7), then analyze the chain for a better direction. |
Some skills require macOS-specific tools:
Rich text mode (/clipboard slack) requires pbcopy-html, a Swift script included in skills/clipboard/:
ln -s "$(pwd)/skills/clipboard/pbcopy-html.swift" /usr/local/bin/pbcopy-html
Plain text mode uses the built-in pbcopy and works out of the box on macOS.
Speech mode requires macOS say and two scripts included in skills/saysay/:
ln -s "$(pwd)/skills/saysay/saysay" /usr/local/bin/saysay
ln -s "$(pwd)/skills/saysay/say-phonetic" /usr/local/bin/say-phonetic
Phonetic mappings are stored per user in ~/.local/share/saysay/phonetics.json (XDG).
Requires Playwright installed globally:
npm install -g playwright
Every plugin's version in packages/<name>/.claude-plugin/plugin.json follows the format 1.0.{commits}, where commits is the number of commits that touched packages/<name>/ or skills/<name>/. Versions bump automatically via the repo's pre-commit hook.
Recovery after rebase or manual edits:
bin/plugin-versions --check # Report drift
bin/plugin-versions --write # Fix drift
Executes bash commands
Hook triggers when Bash tool is used
Modifies files
Hook triggers on file write and edit operations
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 epologee/leclause-skills --plugin dont-do-thatSingle-occupancy lock for a working directory: a vendor-neutral, cross-platform CLI so only one coding agent occupies a directory at a time. claim/release/check on an atomic exclusive-create lockfile keyed by realpath, with pid-based liveness that self-heals a dead holder and respects a live one. No flock, no native dependency. bonsai and the discipline hooks consume it.
Reference for plugin naming, skill resolution, Claude/Codex adapter sync, marketplace aliases, runtime caches, and plugin:skill invocation.
Routes user feedback about agent behavior to the strongest durable target: hooks, skills, project code, or instruction files.
Marketplace-wide utilities for the laicluse-agent-fieldkit marketplace. Currently ships /whats-new (namespaced as /laicluse-agent-fieldkit:whats-new when needed), a one-stop reader for the post-update CHANGELOG section of any installed plugin.
Git workflow skills, repo-aware push policy, and commit/push hooks for agent sessions and direct CLI commits.
Analyze Claude Code agent session transcripts to identify inefficiencies, anti-patterns, repeated mistakes, missing tooling opportunities, and user frustration signals for continuous improvement
Session workflow helpers - knowledge capture, confusion handling, course correction, clipboard copy
YES.md — PUA says NO, YES says YES. 6-layer AI governance: format → trigger → hooks → anti-slack → gates → memory. Makes AI do things RIGHT with encouragement, not pressure. Available in English, 中文, 日本語.
Non-technical progress summaries for Claude Code work (hides diffs/log noise).
Git commit workflow: atomic commits, validation, conventions
Verification-first engineering toolkit for Claude Code. 15 skills across a 5-phase spine (Investigate → Design → Implement → Verify → Ship), 8 specialist agents, an interactive setup wizard. Every skill has rationalizations + evidence requirements. Built for senior ICs and tech leads.