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/pcircle-ai/memesh/claude-mdgit clone --depth 1 https://github.com/PCIRCLE-AI/memeshWrote 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/pcircle-ai/memesh/claude-md)<a href="https://agentmods.dev/instructions/pcircle-ai/memesh/claude-md"><img src="https://agentmods.dev/badge/instructions/pcircle-ai/memesh/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 | $0.02687 | $0.02687 |
| Opus 5 | $0.01344 | $0.01344 |
| Sonnet 5 | $0.00537 | $0.00537 |
| Haiku 4.5 | $0.00269 | $0.00269 |
Grade A, and why
memesh CLAUDE.md scanned grade A with 1 finding 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.
Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
`spawnSync`, so they report **0% while being well tested**. How it starts
The opening of the file, as written. The whole thing — 206 lines — stays where its author put it; the contents beside it link to each section on GitHub.
MeMesh — instructions for AI coding assistants
This file is a pointer, on purpose. It used to carry its own copy of the module tree, the dependency list and the development standards, and a copy is a thing that drifts. It was the last file in the repository still quoting a benchmark figure (95.40% R@5) that release 4.2.11 was spent proving wrong, and its test count was 44 behind. It was also untracked, so no reviewer ever saw it change. Both problems had one cause: it duplicated documents that already exist, are already public, and are already checked by CI.
So — read the real documents. Do not restate them here.
| Question | Read |
|---|---|
| How do I contribute, what must a PR include, which docs move with a code change | CONTRIBUTING.md |
| What are the modules, how does data flow, why is it built this way | docs/ARCHITECTURE.md |
| What is the MCP / HTTP / CLI surface, exactly | docs/api/API_REFERENCE.md |
| What does the product do, how is it installed | README.md |
| Colour, type, spacing, interaction — before ANY dashboard change | DESIGN.md |
| How do I report a vulnerability | SECURITY.md |
| I am an agent INSTALLING memesh for a user | llms-install.md |
| I am an agent USING memesh (the loop, the 11 tools, hygiene) | AGENTS.md |
| What changed, and what is merged but unreleased | CHANGELOG.md ([Unreleased]) |
The few things that live only here
Everything below is either non-obvious from the code or specific to working with an assistant. If anything here starts duplicating a document above, delete it here and link instead.
Running the tests
node scripts/run-tests-isolated.mjs # whole suite, against a throwaway HOME
npm test -- --run # vitest directly — uses YOUR ~/.memesh
Prefer the first. The suite writes to ~/.memesh, so running vitest directly
mutates your real knowledge graph.
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 · 206 lines · 2,687 tokens per session scan A d11d498999aa
memesh CLAUDE.md is an instructions file published in the GitHub repository PCIRCLE-AI/memesh (15 stars, last pushed today), licensed MIT. It adds 2,687 tokens to every session, about $0.0134 per session on Opus 5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
wenlan AGENTS.md
AGENTS.md instructions for 7xuanlu/wenlan, covering agents.md, start here, build and verification, repository invariants and git and release.
wenlan CLAUDE.md
Claude Code instructions for 7xuanlu/wenlan, a project described as: Wenlan is a knowledge base for the AI-native age. Your AI agents capture what they learn, Wenlan keeps it current and distills it into source-cited wiki pages you can trust.
lesson CLAUDE.md
Instructions for OussemaBenAmeur/lesson, covering lesson — ai context, the model, why the concept list exists, evidence signals, weakest to strongest and files.
LockedIn CLAUDE.md
Instructions for daypunk/LockedIn, covering claude.md, what this is, current state, repo layout and working agreements.
rawthink-mcp CLAUDE.md
Claude Code instructions for ygtalp/rawthink-mcp, covering rawthink — system directives, role, tone, thinking directives and epistemic transparency (end of each response).
AI-Prompt-Guide-MCP AGENTS.md
AGENTS.md instructions for Blakeem/AI-Prompt-Guide-MCP, covering ai-prompt-guide mcp server - developer guide, project overview, configuration, zero-config default and optional override.