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/brainqub3/rlm/claude-mdgit clone --depth 1 https://github.com/brainqub3/RLMWrote 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/brainqub3/rlm/claude-md)<a href="https://agentmods.dev/instructions/brainqub3/rlm/claude-md"><img src="https://agentmods.dev/badge/instructions/brainqub3/rlm/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 | $0.01141 | $0.01141 |
| Opus 5 | $0.00571 | $0.00571 |
| Sonnet 5 | $0.00228 | $0.00228 |
| Haiku 4.5 | $0.00114 | $0.00114 |
Grade A, and why
RLM 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 3d 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 — 87 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Project instructions
Headless Codex sub-agent
This repo exposes a root-level headless Codex command for agents that need a non-interactive Codex pass:
./codex-headless "review the current repo and list the highest-risk files"
On Windows shells, use either:
.\codex-headless.ps1 "review the current repo and list the highest-risk files"
.\codex-headless.cmd "review the current repo and list the highest-risk files"
The wrapper runs codex exec from the repo root with gpt-5.5,
model_reasoning_effort="xhigh", --sandbox workspace-write,
approval_policy="never", and --ephemeral. It preserves normal
codex exec behavior, so callers can pass a prompt argument, pipe stdin, add
flags such as --json, or use -o <file> for the final message. It requires
the Codex CLI on PATH and an existing codex login or CODEX_API_KEY scoped
to the invocation.
RLM mode for long-context tasks
This repository includes a faithful "Recursive Language Model" (RLM) setup for Claude Code (after Recursive Language Models, arXiv:2512.24601, Algorithm 1):
- Skill:
rlmin.claude/skills/rlm/ - Persistent Python REPL:
.claude/skills/rlm/scripts/rlm_repl.py— holds the large context as a variable and exposesllm_query/llm_query_map/rlm_queryandFINAL/FINAL_VAR. - Sub-LM (
llm_query): a nested headless Claude Code (claude -p, tools off, default modelhaiku), called programmatically from REPL code — not a Task subagent. The recursiverlm_queryrunsclaude -pwith bash + this skill on.
When the user needs you to work over a context that is too large to paste into chat:
- Ask for (or locate) a context file path.
- Run the
/rlmSkill and follow its procedure.
Keep the main conversation light: the root model never reads the full context —
it writes REPL code that sub-queries the context in chunks, then synthesises.
Use python (not python3) to invoke the REPL on this machine.
OOLONG eval — where run artifacts go
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.
- 3d ago First seen · 87 lines · 1,141 tokens per session scan A c42b5f7f3d3a
RLM CLAUDE.md is an instructions file published in the GitHub repository brainqub3/RLM (393 stars, last pushed 2mo ago), licensed MIT. It adds 1,141 tokens to every session, about $0.0057 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
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
next.js AGENTS.md
Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.