lean-and-mean

Overview → spinlockdevelopment.github.io/lean-and-mean · in-depth guide

A Claude Code and Codex plugin. It writes concise prose and YAGNI code rules into your project's AGENTS.md once, so they apply every session with no plugin code in the loop. /endsession closes a session: it writes what was learned back into that file, commits, pushes, then stops.

Prose. Answer first. Concise, active sentences in plain technical English, broadly following ASD-STE100. No filler or repetition. A diagram or an HTML page when that explains it better.
Code. Map what the change must reach and could break, finish all of it, then a YAGNI ladder: skip speculative work, then reuse the repo, stdlib, the platform, an installed dependency, one line, and only then new code. Never cut input validation, data-loss handling, security, or accessibility.
Memory. Each correction becomes one binding line under ## Rules, so it does not happen twice.

Why use it

Faster, fewer tokens, lower cost: short answers, less code to review, and short sessions that keep each request's context small. The measured numbers: why sessions should be short.

Install

Claude Code (or /plugin marketplace add … and /plugin install … in the app):

claude plugin marketplace add https://github.com/spinlockdevelopment/lean-and-mean
claude plugin install lean-and-mean@lean-and-mean

Restart Claude Code, then run /lean-and-mean:lean-and-mean once per project.

Codex desktop and CLI (run these commands in a terminal):

codex plugin marketplace add https://github.com/spinlockdevelopment/lean-and-mean
codex plugin add lean-and-mean@lean-and-mean

Reopen the desktop app or CLI after installing. Review the plugin's SessionStart hook in desktop Settings → Hooks or CLI /hooks, then start a chat at the project root and run $lean-and-mean once.

To refresh an existing Git marketplace, run codex plugin marketplace upgrade lean-and-mean and reopen Codex.

For the floating Session panel on macOS, build and install Codex Companion from a checkout. The companion uses its own activity hooks and shows estimated cache idle time.

Local checkouts, manual installs, and Codex notes: install details.

Use

Claude Code	Codex	Effect
`/lean-and-mean:lean-and-mean`	`$lean-and-mean`	Create or review `AGENTS.md`: refresh the block, reorder, verify commands and paths, prune, split past 250 lines. Usually runs on its own
`/lean-and-mean:lean-and-mean debt`	`$lean-and-mean debt`	List every `// lean:` shortcut with its upgrade path
`/lean-and-mean:endsession`	`$endsession`	Save Rules and Next, commit, push, and run `## Done` steps when the work is complete. Then stop. Details
Session Status Mod	—	Cache Idle bar (minutes since the last reply) and End session button (a Session pane on desktop with project, branch, context and 5h/weekly limit bars (growth since the last prompt and today in a darker shade, time to the weekly reset), the task list, and a brief Next sessions list from `AGENTS.md`), auto `/endsession` with 5 minutes of cache left, a Rehydrate / Clear first question when the cache has gone cold. Off in `/config`. Details
`@agent-lean-and-mean:explainer`	—	Explainer Agent: local `.pages/<slug>.html` page, written as short Markdown and rendered by the explainer kit

A manual install into ~/.claude/skills/ gives the bare /lean-and-mean and /endsession.

Extras (/lean-and-mean:statusline, /lean-and-mean:typesafe-ai): extras.

How it works

Claude Code loads AGENTS.md through a one-line CLAUDE.md stub (@AGENTS.md); Codex loads it directly. Once the ## Operating Mode block is in it, the mode is on. One SessionStart hook stays silent unless the block is missing or a review is due. Hook details.

Off: disable the plugin and delete the Operating Mode block. Disabling alone leaves the rules active.

It governs code shape, prose, and context files, not correctness. Pair it with /code-review; /simplify reviews code bloat.

Development

python3 -m unittest discover -s tests -v
claude plugin validate .claude-plugin/plugin.json && claude plugin test .

The Codex skill validators reject Claude's argument-hint and disable-model-invocation frontmatter. That is a known exception: the fields stay for Claude, and Codex's explicit-only policy is in skills/endsession/agents/openai.yaml.

License

MIT. See LICENSE.

skills/typesafe-ai/ is copied from typesafe-ai/skills, tracked as the vendor/typesafe-ai-skills submodule. Copyright and credit belong to TypeSafe AI (MIT, Copyright (c) 2026 TypeSafe AI; see skills/typesafe-ai/LICENSE). The only change is routing Jev calls through OpenRouter.