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/e2its/myrmion-graphrag/claude-mdgit clone --depth 1 https://github.com/e2its/myrmion-graphragWrote 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/e2its/myrmion-graphrag/claude-md)<a href="https://agentmods.dev/instructions/e2its/myrmion-graphrag/claude-md"><img src="https://agentmods.dev/badge/instructions/e2its/myrmion-graphrag/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 | $0.01198 | $0.01198 |
| Opus 5 | $0.00599 | $0.00599 |
| Sonnet 5 | $0.00240 | $0.00240 |
| Haiku 4.5 | $0.00120 | $0.00120 |
Grade A, and why
myrmion-graphrag 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 — 82 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
Contexto del proyecto myrmion-graphRAG para Claude Code.
Qué es esto
Dos memorias 100% locales expuestas a Claude Code como dos servidores MCP:
myrmion-graphrag— GraphRAG sobre los documentos del usuario (LightRAG + Ollama).myrmion-codebase— grafo de código: dependencias, impacto, inventario, histórico.
Todo corre en la máquina del usuario; ningún documento ni código sale al exterior.
Herramientas del servidor myrmion-graphrag (documentos)
Cuando el usuario pregunte por algo que probablemente viva en sus documentos indexados (notas, manuales, papers, decisiones) y no esté en el contexto, usa estas herramientas en lugar de suponer:
buscar_conocimiento(consulta, modo="mix", solo_contexto=True, top_k=40)— recupera el contexto crudo (entidades, relaciones, fragmentos) para que tú razones. Primera opción.anadir_documento(texto, descripcion="")— indexa un texto al vuelo (asíncrono).sincronizar_documento(ruta, texto="")— tras editar un documento indexado: reindexa sin duplicar y versionado (no-op si el hash no cambió; delete+insert/upload si cambió).sincronizar_documentos(carpeta="")— sincroniza una carpeta entera por hash (added/ modified/removed) y crea snapshot.historico_documento(ruta)/estado_documentos()— histórico y estado del versionado.estado_rag()— salud de LightRAG + backend de storage activo. Llámalo primero si algo falla.verificar_alineacion()/reconciliar(aplicar=False)— solo perfil híbrido: comprueba y repara la alineación Neo4j⇄Postgres.
Modos: mix (recomendado), hybrid, local (entidades), global (temas), naive (vectores).
Herramientas del servidor myrmion-codebase (código)
Para preguntas sobre el código del usuario (impacto, dependencias, qué es reutilizable/muerto):
indexar_codebase(ruta="", incremental=False)— indexa/reindexa.dependencias_de(simbolo, profundidad=1)— de qué depende.quien_llama_a(simbolo, profundidad=1)— quién lo llama.a_que_afecta(simbolo, profundidad=5)— blast radius si se cambia.inventario(filtro=""),codigo_muerto(),arquitectura()— inventario y visión global.cambios_desde(git_ref),historico(simbolo),estado_indexado().anotar_simbolo(simbolo, etiqueta, nota="")— fija mandatory/reusable/keep/deprecated.sincronizar_codigo(rutas, durable=False)— sync incremental tras editar.
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 · 82 lines · 1,198 tokens per session scan A 9ad480cebd72
myrmion-graphrag CLAUDE.md is an instructions file published in the GitHub repository e2its/myrmion-graphrag (0 stars, last pushed 2mo ago), licensed MIT. It adds 1,198 tokens to every session, about $0.0060 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
repo-graphrag-mcp AGENTS.md
Instructions for yumeiriowl/repo-graphrag-mcp, covering repo graphrag mcp usage guide, prerequisites, three tools, tool 1: graphcreate and tool 2: graphquery.
James-RAG-Evol CLAUDE.md
Claude Code instructions for Hashevolution/James-RAG-Evol, covering sekos / james — session briefing for claude code, what james is, where we are right now, critical rules for this session and where to look next.
HyperHierarchicalRAG copilot-instructions.md
Instructions for u9401066/HyperHierarchicalRAG, covering copilot instructions for zotero + pubmed mcp, 可用的 mcp servers, 1. zotero keeper, 2. pubmed search and 核心工作流程.
agentic-graph-rag-skills AGENTS.md
Instructions for AnthonyAlcaraz/agentic-graph-rag-skills, covering agents.md - how an ai should read this repository and for ai agents reading this.
HyperHierarchicalRAG CLAUDE.md
Instructions for u9401066/HyperHierarchicalRAG, covering claude.md - claude code 專案指引, 專案概述, 法規層級, 核心原則 and 0. 開發哲學 💡.
rag-rat AGENTS.md
Instructions for cq27-dev/rag-rat, covering agents.md, prefer the rag-rat mcp for code browsing, record durable learnings as rag-rat memories, repo orientation and worktree correctness.