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/vvooki-sys/loomem/claude-mdgit clone --depth 1 https://github.com/vvooki-sys/loomemWrote 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/vvooki-sys/loomem/claude-md)<a href="https://agentmods.dev/instructions/vvooki-sys/loomem/claude-md"><img src="https://agentmods.dev/badge/instructions/vvooki-sys/loomem/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 | $0.03313 | $0.03313 |
| Opus 5 | $0.01656 | $0.01656 |
| Sonnet 5 | $0.00663 | $0.00663 |
| Haiku 4.5 | $0.00331 | $0.00331 |
Grade A, and why
loomem 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 4d 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 — 122 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md — rules for AI code generators in the Loomem repo
Hard rules every AI assistant (Claude Code, Cowork, Copilot, Cursor, …) must follow when touching this repo. Some are enforced by CI; the rest are enforced in review. Deep rationale, complexity carve-outs and edge cases live in docs/agent-conventions.md — this file stays short on purpose, because it loads into context every session.
What Loomem is. An open-core, single-user memory engine (Apache-2.0, vvooki-sys/loomem), forked from Engram. The repo is public — assume external readers and their agents. Public positioning: "the open-source context layer for LLM agents." It is memory-only (the memory_* MCP tool set, no file layer) and single-user (no accounts, RBAC, SSO, multi-tenancy). Those two product boundaries are frozen (see § Product boundaries).
Router — read the subset that applies
- Changing Rust code → § Five hard rules + § Build & run + § Repo map + § Rust rules.
- Docs / landing / content / GEO → § Docs & site + hard rule 5 (surgical). A docs task never touches Rust.
- Cutting a release → § Releases.
- Always → § Five hard rules, especially live-state (rule 4), product boundaries (rule 3), secrets (rule 1).
Five hard rules (never break these)
-
Secrets & personal data never get committed. No API keys, tokens, passwords, production endpoints — and no real names or private data in
entities.toml/synonyms.toml(gitignored; only*.examplevariants are tracked). Any user content that gets persisted anywhere, including derived/audit fields likeoriginal_content, must pass throughsanitizer+ the PII filter — not just the visiblecontent..gitleaks.tomlcatches some leaks; you are the first gate. See something sensitive in a diff → stop and tell the user. -
The local gate must be green.
cargo fmt --check && cargo clippy --workspace -- -D warningsis byte-identical to CI and to the pre-commit hook. If it isn't green, the work isn't done. Never use--no-verify.
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.
- 4d ago First seen · 122 lines · 3,313 tokens per session scan A 21ad829bc950
loomem CLAUDE.md is an instructions file published in the GitHub repository vvooki-sys/loomem (4 stars, last pushed 5d ago), licensed Apache-2.0. It adds 3,313 tokens to every session, about $0.0166 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 2026-08-31.
Other instructions, from other repositories
gno AGENTS.md
AGENTS.md instructions for gmickel/gno, covering knowledge cutoff warning, apis - bun first!, must use bun, acceptable node:\ (no bun equivalent) and testing.
palace-rs AGENTS.md
AGENTS.md instructions for AncientiCe/palace-rs, covering agent rules, 1. test-driven development (tdd), 2. quality gates on every task, 3. no plan markdown files and 4. internal api discipline.
gno CLAUDE.md
Claude Code instructions for gmickel/gno, a project described as: Local AI-powered document search and editing with first-in-class hybrid retrieval, LLM answers, WebUI, REST API and MCP support for AI clients.
palace-rs CLAUDE.md
Claude Code instructions for AncientiCe/palace-rs, covering claude notes, rtk (rust token killer) - token-optimized commands, golden rule, ❌ wrong and ✅ correct.
gini-agent AGENTS.md
AGENTS.md instructions for Open-Curiosity/gini-agent, covering gini agent instructions, shape, adrs, boundaries and branches.
plur CLAUDE.md
Claude Code instructions for plur-ai/plur, covering claude.md, what is plur, development, package dependency and version bumps.