attestor CLAUDE.md

attestor CLAUDE.md is an instructions file for coding agents from bolnet/attestor. It costs 2,293 tokens per session, scanned A, original, MIT.

Project instructions for Attestor, a self-hosted memory layer for agent teams that retrieves stored information without using a language model in its core retrieval path.

In plain words
What is it for?
Use them when developing or operating Attestor's Python package, command-line interface, API, memory client, and PostgreSQL, vector, or graph storage components.
Why use it?
They give contributors a shared reference for the project's structure, setup, architecture, and supported storage backends.

Instructions file

Install

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.

agentmods
npx agentmods add instructions/bolnet/attestor/claude-md
Clone the repo
git clone --depth 1 https://github.com/bolnet/attestor

Wrote 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.

agentmods badge for attestor CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/bolnet/attestor/claude-md.svg)](https://agentmods.dev/instructions/bolnet/attestor/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/bolnet/attestor/claude-md"><img src="https://agentmods.dev/badge/instructions/bolnet/attestor/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 2,293 This file is loaded in full into every session.
When invoked 2,293 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.02293 $0.02293
Opus 5 $0.01146 $0.01146
Sonnet 5 $0.00459 $0.00459
Haiku 4.5 $0.00229 $0.00229

Measured 5d ago against content hash 307422f7ab2b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

attestor 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 5d 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.

CLAUDE.md · 126 lines

How it starts

The opening of the file, as written. The whole thing — 126 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Attestor

The memory layer for agent teams. Self-hosted, deterministic retrieval, zero LLM in the critical path.

PyPI: attestor -- Python import: attestor.

Project Structure

attestor/
  core.py              -- AgentMemory class (main public API)
  client.py            -- MemoryClient (HTTP drop-in for remote Attestor)
  context.py           -- AgentContext (identity, role, namespace, token budget, provenance)
  models.py            -- Memory, RetrievalResult dataclasses
  cli.py               -- CLI entry point
  api.py               -- Starlette ASGI REST API
  store/
    base.py            -- DocumentStore / VectorStore / GraphStore interfaces
    registry.py        -- Backend selection; default = postgres + neo4j
    connection.py      -- Config layering / env resolution
    embeddings.py      -- Provider auto-detect (local / OpenAI / Bedrock / Vertex / Azure)
    postgres_backend.py-- Postgres (document role; pgvector embedding column kept in schema)
    pinecone_backend.py-- Pinecone (vector role; auto-detects Local Docker vs Cloud)
    neo4j_backend.py   -- Neo4j + GDS (graph role, PageRank / BFS)
    schema.sql
  graph/
    extractor.py       -- Entity/relation extraction (4-output GraphRAG)
  retrieval/
    orchestrator.py    -- 6-step semantic-first cascade (vector → BM25 → RRF → graph → MMR → fit)
    tag_matcher.py
    scorer.py          -- Temporal + entity + PageRank boosts, confidence decay, MMR
  temporal/
    manager.py         -- Contradiction detection, supersession, timeline, as_of replay
  extraction/          -- Rule-based + optional LLM memory extraction
  mcp/
    server.py          -- MCP server (tools + resources + prompts)
  hooks/
    session_start.py   -- Context injection at session start
    post_tool_use.py   -- Auto-capture from Write/Edit/Bash
    stop.py            -- Session summary on exit
  utils/
    config.py, tokens.py
  infra/local/         -- Docker Compose for local Postgres + Pinecone Local + Neo4j
tests/                 -- Unit tests (live integration tests env-gated)

Read the full file on GitHub · 126 lines

Changes

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.

  1. 5d ago First seen · 126 lines · 2,293 tokens per session scan A 307422f7ab2b

Subscribe to this mod's changes

attestor CLAUDE.md is an instructions file published in the GitHub repository bolnet/attestor (14 stars, last pushed 3mo ago), licensed MIT. It adds 2,293 tokens to every session, about $0.0115 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.