Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/ofux/lauren/claude-mdgit clone --depth 1 https://github.com/ofux/laurenWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/ofux/lauren/claude-md)<a href="https://agentmods.dev/instructions/ofux/lauren/claude-md"><img src="https://agentmods.dev/badge/instructions/ofux/lauren/claude-md.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.03213 | $0.03213 |
| Opus 5 | $0.01606 | $0.01606 |
| Sonnet 5 | $0.00643 | $0.00643 |
| Haiku 4.5 | $0.00321 | $0.00321 |
Grade A, and why
lauren CLAUDE.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 113 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
Commands
npm run build # tsc → dist/, then chmod +x the lauren bin
npm run watch # tsc --watch
npm run clean # remove dist/
npm run check # biome check --write (lint + format + organize-imports)
npm run lint # biome lint only
npm run format # biome format --write
npm test # vitest run (one-shot)
npm run test:watch # vitest in watch mode
Tests live next to source as *.test.ts (Vitest, ESM, Node >=20). Verify changes by npm run build && npm run check && npm test. Manual smoke test: from a throwaway repo, run lauren vibe --dry-run to print the queue, and lauren --list to inspect state (plain table; omit --list for the interactive TUI).
External binaries
The runtime shells out to two CLIs that must be on $PATH:
claude— used interactively (planning, spec) and as the default agent for the implement / fix pipeline phases, brain JSON decisions, and merger conflict resolution.codex— used as the default agent for the review pipeline phase (codex exec review -o <file> <prompt>).
Per-phase agent selection is configurable in .lauren/config.json under
agents: each of implement, review, fix, merger, and brain
independently accepts "claude" or "codex". The adapters live in
src/agents/ and implement the CodingAgent port (runEdit, runReview,
runJson) — see src/agents/types.ts.
If a binary is missing for any role configured to use it, lauren vibe will
fail at the corresponding step.
Architecture
Lauren is a single-daemon system that drains a plan queue end-to-end without further human input.
lauren CLI (src/bin/lauren.ts) — the single public executable. lauren plan spawns interactive claude with PLAN_SYSTEM_PROMPT; that claude session writes .lauren/plans/<slug>.md and re-invokes the CLI as lauren _register <slug> --path ... --title ... (a hidden subcommand) to add a row with status enqueued to .lauren/plans.json. The default lauren (no subcommand) is an interactive Ink TUI (src/tui/TodoApp.tsx) showing the queue; --list (or non-TTY) prints a static colored table instead. Selecting a row dispatches to cancelPlan (src/cancel.ts) which routes by status (see Cancellation below). The TUI also supports two queue-level actions: t on a failed row resets it to ready via retryPlan (src/retry.ts), and r triggers a brain reorganize pass that re-thinks the whole ready queue (refused while lauren vibe is running, detected via .lauren/vibe.pid).
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 5d ago First seen · 113 lines · 3,213 tokens per session scan A c4fd5a934ba0
lauren CLAUDE.md is an instructions file published in the GitHub repository ofux/lauren (12 stars, last pushed 3mo ago), licensed MIT. It adds 3,213 tokens to every session, about $0.0161 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
SeekMoney-ai CLAUDE.md
Claude Code instructions for zykooooooooo/SeekMoney-ai, covering claude.md, project overview, development commands, development and production.
spine CLAUDE.md
Claude Code instructions for synaptixs/spine, covering spine — working on this repo, layout — where things live, invariants — break these and things get subtly wrong, gotchas that have bitten and before pushing.
t-0-spec-kit-ralph AGENTS.md
Instructions for T-0-co/t-0-spec-kit-ralph: T-0's upstream repository for the speckit-ralph autonomous task execution framework. Installer and library for spec-kit-driven development with Ralph Wiggum Loop.
open-responses-agent-skill AGENTS.md
Instructions for OthmanAdi/open-responses-agent-skill, covering open responses agent development, about this skill, core knowledge, single unified endpoint and provider selection via model suffix.
loopforge AGENTS.md
AGENTS.md instructions for taberoajorge/loopforge, covering loopforge, stack, commands, architecture and conventions.
loopforge CLAUDE.md
Claude Code instructions for taberoajorge/loopforge, covering loopforge, quick start, architecture, critical rules and source of truth.