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/rtk-ai/icm/claude-mdgit clone --depth 1 https://github.com/rtk-ai/icmWrote 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/rtk-ai/icm/claude-md)<a href="https://agentmods.dev/instructions/rtk-ai/icm/claude-md"><img src="https://agentmods.dev/badge/instructions/rtk-ai/icm/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 | $0.05841 | $0.05841 |
| Opus 5 | $0.02920 | $0.02920 |
| Sonnet 5 | $0.01168 | $0.01168 |
| Haiku 4.5 | $0.00584 | $0.00584 |
Grade A, and why
icm 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 — 761 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Projet: ICM (Infinite Context Memory)
Un système de mémoire persistante pour LLMs, écrit en Rust, avec support MCP pour intégration Claude Code.
Objectif
Créer une mémoire long-terme intelligente qui:
- Stocke des memories avec embeddings dans Turso/SQLite
- Fait du retrieval hybride (BM25 + vector similarity)
- Gère le decay temporel et la consolidation
- Expose un serveur MCP pour que Claude Code puisse stocker/récupérer des souvenirs
Stack technique
- Rust (edition 2021)
- Turso/libsql pour le storage
- sqlite-vec pour les embeddings vectoriels
- tokio pour l'async
- serde pour la serialization
- clap pour le CLI
- MCP SDK Rust (ou implémentation manuelle du protocole JSON-RPC)
Structure du workspace
icm/
├── Cargo.toml (workspace)
├── crates/
│ ├── icm-core/ # Types et traits fondamentaux
│ │ ├── Cargo.toml
│ │ └── src/
│ │ ├── lib.rs
│ │ ├── memory.rs # Struct Memory, Importance, MemorySource
│ │ ├── store.rs # Trait MemoryStore
│ │ └── embedder.rs # Trait Embedder
│ │
│ ├── icm-store-turso/ # Implémentation Turso du store
│ │ ├── Cargo.toml
│ │ └── src/
│ │ ├── lib.rs
│ │ ├── turso.rs # TursoStore impl MemoryStore
│ │ └── migrations.rs # Schema SQL
│ │
│ ├── icm-retriever/ # Recherche hybride
│ │ ├── Cargo.toml
│ │ └── src/
│ │ ├── lib.rs
│ │ ├── hybrid.rs # HybridRetriever (BM25 + vector)
│ │ └── reranker.rs # Reranking optionnel
│ │
│ ├── icm-extractor/ # Extraction de memories depuis du texte
│ │ ├── Cargo.toml
│ │ └── src/
│ │ ├── lib.rs
│ │ └── llm.rs # Appel LLM pour extraire facts/summaries
│ │
│ ├── icm-mcp/ # Serveur MCP pour Claude Code
│ │ ├── Cargo.toml
│ │ └── src/
│ │ ├── main.rs
│ │ ├── tools.rs # icm_store, icm_recall, icm_forget
│ │ └── protocol.rs # JSON-RPC MCP handling
│ │
│ └── icm-cli/ # CLI standalone
│ ├── Cargo.toml
│ └── src/
│ └── main.rs # Commands: store, recall, consolidate, prune
│
├── config/
│ └── default.toml # Config par défaut
│
└── README.md
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 · 761 lines · 5,841 tokens per session scan A 75a5a7d4de3a
icm CLAUDE.md is an instructions file published in the GitHub repository rtk-ai/icm (552 stars, last pushed 5d ago), licensed Apache-2.0. It adds 5,841 tokens to every session, about $0.0292 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
grit AGENTS.md
Instructions for rtk-ai/grit, covering persistent memory (icm), recall (before starting work), store (after completing significant work) and other commands.
grit CLAUDE.md
Instructions for rtk-ai/grit, covering persistent memory (icm), recall (before starting work), store (after completing significant work), other commands and rtk (rust token killer) - token-optimized commands.
vespertide AGENTS.md
AGENTS.md instructions for dev-five-git/vespertide, covering vespertide knowledge base, structure, where to look, data flow and conventions.
aeon CLAUDE.md
Instructions for aeonfun/aeon, covering aeon, how aeon works, strategy, voice and soul file hierarchy (read in this order).
mcp-structured-memory CLAUDE.md
Claude Code instructions for nmeierpolys/mcp-structured-memory, a project described as: Structured Memory MCP Server.
inkwell-memory CLAUDE.md
Instructions for veronchenko/inkwell-memory, covering claude.md — inkwellmemory, layout, multi-tenant mode (inkwellmultitenant=1), conventions and testing.