Borrowing it
Nothing to install: this file belongs to lossless-claude/lcm. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/lossless-claude/lcm/main/AGENTS.mdgit clone --depth 1 https://github.com/lossless-claude/lcmWrote 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/lossless-claude/lcm/agents-md)<a href="https://agentmods.dev/instructions/lossless-claude/lcm/agents-md"><img src="https://agentmods.dev/badge/instructions/lossless-claude/lcm/agents-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/lossless-claude/lcm/agents-md"><img src="https://agentmods.dev/badge/instructions/lossless-claude/lcm/agents-md.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.01005 | $0.01005 |
| Opus 5.5 | $0.00402 | $0.00402 |
| Sonnet 5.5 | $0.00201 | $0.00201 |
| Haiku 4.5 | $0.00101 | $0.00101 |
Grade A, and why
lcm AGENTS.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 1mo 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 — 50 lines — stays where its author put it; the contents beside it link to each section on GitHub.
lcm — agent contract
This file routes; it does not repeat what the sources say. Read CONTEXT.md first: it fixes the vocabulary.
Sources of truth
| Working on | Read | The code it must agree with |
|---|---|---|
| anything | CONTEXT.md | src/transcript.ts, src/subagent-attribution.ts |
| the data model, compaction, ingestion | docs/architecture.md | src/store/, src/capture.ts, src/compaction.ts, src/daemon/ |
| hooks, the function-hooks module | docs/hook-protocol.md | .claude-plugin/plugin.json, src/hooks/, hooks/lcm-hooks.ts |
env vars, config.json, the summarizer |
docs/configuration.md | src/db/config.ts, src/daemon/config.ts |
| MCP tools and their parameters | docs/agent-tools.md | src/mcp/tool-catalog.ts |
| cutting a release | docs/releasing.md | .github/workflows/publish.yml, .github/workflows/version-pr.yml |
| CI, the self-hosted runner | docs/ci-runner.md | .github/workflows/ci.yml |
A change that alters what one of these documents states updates the document in the same PR.
What is the plugin and what is the repository
The plugin is distributed from this repository's root, so everything the plugin loads must be usable by someone who installed lcm and never opens this repo:
- Plugin:
.claude-plugin/plugin.json,skills/,hooks/,bundle/(built bynpm run build:bundle, committed by version PRs). - Repository only:
.claude/,.agents/,.github/,scripts/,test/,tools/,plans/, and every doc about developing lcm rather than using it.
Development skills have two homes and no copies: .agents/skills/ for what Codex and Copilot load (both read that directory), .claude/skills/ for what only Claude Code loads.
A skill that exists to work on lcm goes under .claude/ or .agents/, never under the plugin. The plugin's own surface is skills only: a skill that only a person should invoke sets disable-model-invocation: true, and it routes to lcm help <command> rather than restating options, so the CLI stays the one reference.
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.
- 1mo ago Changed · +1 tokens per session 4c9ff9475d3e
- 1mo ago Changed · +20 tokens per session e797b26bb114
- 1mo ago Changed · +6 tokens per session 006505f28411
- 1mo ago Changed · +4 tokens per session 07985abf8248
- 1mo ago Changed · +49 lines e3352bfd0146
- 1mo ago Changed · -183 lines e3b0c44298fc
- 2mo ago First seen · 184 lines · 2,386 tokens per session scan C 4f37e517e772
lcm AGENTS.md is an instructions file published in the GitHub repository lossless-claude/lcm (60 stars, last pushed 4d ago), licensed MIT. It adds 1,005 tokens to every session, about $0.0040 per session on Opus 5.5. 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).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
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 copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
deepseek-harness AGENTS.md
AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.