By epologee
Dispatch a rover at a task and stay back while it decides in the field: a SURVEY/DRIVE/INSPECT/STOW phase machine with a decide framework, pride and trim quality gates, verify evidence discipline, prepare for queuing missions across repos, and a stop communique. Asks the autonomous layer at startup whether it needs to keep itself alive. Depends on gurus for the INSPECT panel review.
Choice framework for a rover in the field, or for any moment an operator is stuck between options. Classifies the call, resolves it, and logs the verdict when a loop file is present.
Lay a rover loop file in another repo's .autonomous/ now, so the operator can wake it later from inside that repo. Triggers on /rover:prepare. Writes the loop file and .gitignore with full mission context, but does not start host continuation, create the mission branch, or run SURVEY. The operator drives those at wake time.
Pride check. Spawns a contrarian agent that reviews an artefact with a skeptical eye and surfaces what the operator would hate but the rover missed. Hard gate on every artefact the rover produces, not just pushes.
Rover briefing. Explains what the autonomous Rover does and how to dispatch, steer, and stop one. Read this when you are about to send a Rover out for the first time, or when you forgot which command does what.
End a rover mission on purpose. Stops or marks the host continuation, writes a final log entry, and transmits a mission report home: a length-scaled narrative of the traverse, a qualitative conclusion, a (usually empty) not-done listing, and any external-action gates the rover could not take itself (push, merge, deploy). Reached via the rover entry point.
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.
Agent tooling for Claude Code and Codex.
The marketplace currently ships:
how-plugins-work: reference material for plugin names, skill names,
marketplace aliases, manifests, adapters, and runtime caches.self-improvement: routes feedback about agent behavior to hooks, skills,
project code, or instruction files.git-discipline: git workflow skills plus commit and push hooks for agent
sessions and direct CLI commits.dont-do-that: guardrail hooks for Claude Code plus /duh and
/just-a-question correction skills for Claude Code and Codex.intervision: bring another coding agent in as a peer to review work just
done or just discussed. Claude hands work to Codex; Codex hands work to
Claude.anger-management: curse at your coding agent now, fix the real problem
later. Capture commands log friction to a global pile; a delayed background
investigation diagnoses the pattern and repair routes the fix.autonomous (Claude Code): keep an autonomous mission running across turns. A startup
capability probe decides whether keep-alive machinery (cron heartbeat,
backoff, wake) is needed; persistent processes run without it.gurus: opinionated review panels for code, decisions, and prose. The
orchestrator routes to the software, council, or writers panel.rover: dispatch a rover at a task and stay back while it decides in the
field: a phase machine with decide, pride/trim quality gates, verify
evidence discipline, and a stop communique.clipboard: copy the core content of the last answer to the macOS
clipboard. Plain text by default, /clipboard slack for rich text.laicluse-agent-fieldkit (Claude Code): marketplace-wide utilities.
/whats-new [plugin]
re-reads the latest CHANGELOG section of any installed plugin, or the
marketplace-wide news without arguments. Use
/laicluse-agent-fieldkit:whats-new only when a namespaced form is needed.Claude Code:
claude plugins marketplace add epologee/laicluse-agent-fieldkit
claude plugins install how-plugins-work@laicluse-agent-fieldkit
claude plugins install self-improvement@laicluse-agent-fieldkit
claude plugins install git-discipline@laicluse-agent-fieldkit
claude plugins install dont-do-that@laicluse-agent-fieldkit
claude plugins install intervision@laicluse-agent-fieldkit
claude plugins install anger-management@laicluse-agent-fieldkit
claude plugins install autonomous@laicluse-agent-fieldkit
claude plugins install gurus@laicluse-agent-fieldkit
claude plugins install rover@laicluse-agent-fieldkit
claude plugins install clipboard@laicluse-agent-fieldkit
claude plugins install laicluse-agent-fieldkit@laicluse-agent-fieldkit
Codex:
codex plugin marketplace add epologee/laicluse-agent-fieldkit
codex plugin add how-plugins-work@laicluse-agent-fieldkit
codex plugin add self-improvement@laicluse-agent-fieldkit
codex plugin add git-discipline@laicluse-agent-fieldkit
codex plugin add dont-do-that@laicluse-agent-fieldkit
codex plugin add intervision@laicluse-agent-fieldkit
codex plugin add anger-management@laicluse-agent-fieldkit
codex plugin add gurus@laicluse-agent-fieldkit
codex plugin add rover@laicluse-agent-fieldkit
codex plugin add clipboard@laicluse-agent-fieldkit
Codex receives the generated subset that has Codex-compatible runtime behavior. Claude-only plugins and skills are intentionally omitted from the Codex marketplace instead of being served as inert commands.
Existing installs under the previous marketplace identity do not rename themselves. See docs/migration.md for the hard-cutover commands.
For local development, point the marketplace at this working copy:
claude plugins marketplace add ./
codex plugin marketplace add ./
Claude metadata is the source. Codex metadata is generated:
bin/plugin-adapters build .
bin/plugin-adapters check .
bin/plugin-adapters diff .
Plugin versions follow 2.0.<commit-count> per package:
bin/plugin-versions --check
bin/plugin-versions --write
Enable the local git hooks in this clone:
git config core.hooksPath hooks
The pre-commit hook bumps versions, builds Codex adapters, and stages the
generated targets. The commit-msg hook requires PII-Doublecheck: yes.
All first-party runtime state for l'Aicluse Agent Fieldkit uses:
${LAICLUSE_HOME:-$HOME/.laicluse}
Agent-harness caches stay where the harness expects them, for example
~/.claude/plugins/cache and ~/.codex/plugins/cache.
npx claudepluginhub epologee/laicluse-agent-fieldkit --plugin roverGuardrail 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.
Single-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.
Git workflow skills, repo-aware push policy, and commit/push hooks for agent sessions and direct CLI commits.
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Develop, test, build, and deploy Godot 4.x games with Claude Code. Includes GdUnit4 testing, web/desktop exports, CI/CD pipelines, and deployment to Vercel/GitHub Pages/itch.io.
A growing collection of Claude-compatible academic workflow bundles. Covers scientific figures, manuscript writing and polishing, reviewer assessment, citation retrieval, data availability, paper reading, literature search, response letters, paper-to-PPTX conversion, and evidence-grounded Chinese invention patent drafting. Rules are organized as reusable skill folders with explicit workflows and quality checks.
Comprehensive PR review agents specializing in comments, tests, error handling, type design, code quality, and code simplification
Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review
UI/UX design intelligence. 67 styles, 161 palettes, 57 font pairings, 25 charts, 15 stacks (React, Next.js, Vue, Svelte, Astro, SwiftUI, React Native, Flutter, Tailwind, shadcn/ui, Nuxt, Jetpack Compose). Actions: plan, build, create, design, implement, review, fix, improve, optimize, enhance, refactor, check UI/UX code. Projects: website, landing page, dashboard, admin panel, e-commerce, SaaS, portfolio, blog, mobile app. Elements: button, modal, navbar, sidebar, card, table, form, chart. Styles: glassmorphism, claymorphism, minimalism, brutalism, neumorphism, bento grid, dark mode, responsive, skeuomorphism, flat design. Topics: color palette, accessibility, animation, layout, typography, font pairing, spacing, hover, shadow, gradient.