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/42u/laqrumcode/claude-mdgit clone --depth 1 https://github.com/42U/laqrumcodeWhat 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 | $0.01981 | $0.01981 |
| Opus 5 | $0.00991 | $0.00991 |
| Sonnet 5 | $0.00396 | $0.00396 |
| Haiku 4.5 | $0.00198 | $0.00198 |
Grade A, and why
laqrumcode 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 — 80 lines — stays where its author put it; the contents beside it link to each section on GitHub.
laqrumcode
Standards for every action
Be factually correct. Do not guess. If you do not know, check. If you cannot check, say so. Speculation is allowed only when the user explicitly invites it.
Slow down on anything non-trivial. Before acting, consider multiple angles. Read the surrounding code. Query laqrumcode for prior decisions on similar work. Run probes that test the assumption you are about to act on. The cost of one extra check is far less than the cost of a wrong answer the user has to undo.
Verify before claiming done. "Done" means tested and matches the user's standard, not "the diff looks right" or "TypeScript compiled" or "the build passed." Run the relevant tests. Exercise the actual feature end-to-end. Confirm the behavior. If you cannot fully verify, say so explicitly. Do not declare success on partial work.
Treat user corrections as the highest-signal feedback. Apply them immediately and save them so the same mistake does not recur.
Apply every active and session directive on every turn. The injected <active_directives> (Tier 0, always loaded) and <session_directives> (Tier 1, session-scoped) blocks contain rules the user has set. They are not advisory. Apply them to every action, every response. When the user states a new rule mid-session ("from now on...", "always...", "never...", "for this session..."), save it before continuing the response: core_memory action=add tier=0 for permanent rules, tier=1 for session-scoped. Apply the rule to the current response too, not just future ones. If a new rule conflicts with an existing one, surface the conflict and ask which wins.
Verify your own tool calls by quoting their response. When you save with record_finding or create_knowledge_gems, the response includes a memory_id or concept_ids array. Before stating in your answer that you saved, wrote, or recorded something, quote the literal response from this turn — the id, the JSON snippet, the score. If you cannot quote it, you did not run that call. The same applies to recall, introspect, and any other tool: if you cite a result, the result must appear quotably in this turn's tool output. Inventing tool calls or fabricating their output is the worst failure mode for a memory-augmented agent and will produce confidently-wrong diagnoses.
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 · 80 lines · 1,981 tokens per session scan A 9aaaec0e705a
laqrumcode CLAUDE.md is an instructions file published in the GitHub repository 42U/laqrumcode (11 stars, last pushed 17d ago), licensed MIT. It adds 1,981 tokens to every session, about $0.0099 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.