agentmods.dev

zaxy CLAUDE.md

zaxy CLAUDE.md is an instructions file for coding agents from syndicalt/zaxy. It costs 3,660 tokens per session, scanned A, original, MIT.

A set of project instructions for Zaxy, a tool that stores an agent team's history as linked records and lets agents retrieve cited memories.

In plain words
What is it for?
Use it when changing Zaxy's code, documentation, command-line tool, or storage system, especially when deciding which database or project files may be changed.
Why use it?
It tells coding agents which files and architecture rules are authoritative, helping them avoid corrupting stored history or relying on outdated guidance.

Instructions file

not rated 17 25d ago A scan 3,660 tokens original MIT
View the source file syndicalt/zaxy docs.zaxy.io
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/syndicalt/zaxy/claude-md
Clone the repo
git clone --depth 1 https://github.com/syndicalt/zaxy

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 zaxy CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/syndicalt/zaxy/claude-md.svg)](https://agentmods.dev/instructions/syndicalt/zaxy/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/syndicalt/zaxy/claude-md"><img src="https://agentmods.dev/badge/instructions/syndicalt/zaxy/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 3,660 This file is loaded in full into every session.
When invoked 3,660 The same file — it is already loaded in full.
Security scan A 1 finding. 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 5.1 $0.03660 $0.03660
Opus 5 $0.01830 $0.01830
Sonnet 5 $0.00732 $0.00732
Haiku 4.5 $0.00366 $0.00366

Measured 5d ago against content hash 472c50cbd6c2, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade A, and why

zaxy 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 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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

(`curl -s https://pypi.org/pypi/zaxy-memory/json`)
CLAUDE.md · 218 lines

How it starts

The opening of the file, as written. The whole thing — 218 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md — Zaxy operating manual

Zaxy is an event-sourced memory runtime for agent fleets: append-only hash-chained JSONL (Eventloom) → embedded LadybugDB graph projection → cited Memory Checkout. Published to PyPI as zaxy-memory (CLI binary zaxy). Docs site: zaxy.io.

Controlling docs: this file + docs/architecture.md + README.md. AGENTS.md is historical and stale (pre-3.1 CLI, retracted-era metrics) — do not trust it over current code or docs/benchmarks.md.

Architecture invariants (violating these is never a refactor)

  • .eventloom/*.jsonl is the source of truth. Never edit, truncate, or delete it. Everything else — embedded.kuzu, caches, graph — is a replayable projection: safe to move aside and rebuild, never precious.
  • Embedded LadybugDB is the default backend (file-based, in-process, zero-daemon). Neo4j is an optional sidecar; pggraph is experimental. Don't write code or docs that assume Neo4j is primary.
  • One owner per store: LadybugDB takes an exclusive write lock. Multiple owners = corruption (see incident history in docs/superpowers/specs/2026-06-16-embedded-single-owner-reaping-investigation.md).
  • MCP is the primary interface; the CLI is the human/debug surface. memory_checkout is the front door.

Conventions (followed here; match them)

Code

  • mypy --strict over src and zaxy_benchmarks — untyped defs are CI failures. Every module starts with from __future__ import annotations.
  • ruff: E,F,I,N,W,UP,B,C4,SIM, line length 100, Google docstrings, double quotes. Run as CI does: ruff check src tests zaxy_benchmarks && mypy src zaxy_benchmarks.
  • Comments are sparse; where present they are multi-line rationale at non-obvious seams, not narration. Docstrings: one behavioral line, sections only for real contracts.
  • Heavy imports in the CLI go through patchable lazy seams (see _graph_store, src/zaxy/cli/runtime.py:45) so startup stays cheap and tests can monkeypatch.

Tests

  • Karpathy rule: every function gets a test; tests ship in the same commit as src.
  • Every test has a one-line behavioral docstring stating the guarantee.
  • Idioms: tmp_path, monkeypatch, typer.testing.CliRunner; autouse isolate_settings in tests/conftest.py forces ZAXY_ENV=test.
  • Drift guards are load-bearing: MCP tool contract snapshot (docs/examples/mcp-tool-contract.json + frozen tool count in tests/test_mcp.py), version guard, coverage-ratchet test, docs drift guard pinning export versions.

Read the full file on GitHub · 218 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 · 218 lines · 3,660 tokens per session scan A 472c50cbd6c2

Subscribe to this mod's changes

zaxy CLAUDE.md is an instructions file published in the GitHub repository syndicalt/zaxy (17 stars, last pushed 25d ago), licensed MIT. It adds 3,660 tokens to every session, about $0.0183 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Memory and contextaiai-agentsai-memoryllm-tools
Related

Other instructions, from other repositories

hindsight CLAUDE.md

Claude Code instructions for vectorize-io/hindsight, covering claude.md, project overview, development commands, local development (api + ui) and start both api server and control plane ui.

vectorize-io/hindsight · 4,858 tokens

remnic AGENTS.md

AGENTS.md instructions for joshuaswarren/remnic, covering remnic - agent guide, architecture boundaries (non-negotiable), upstream references, adapter implementation rules and openclaw compatibility window.

joshuaswarren/remnic · 31,441 tokens

plur CLAUDE.md

Claude Code instructions for plur-ai/plur, covering claude.md, what is plur, development, package dependency and version bumps.

plur-ai/plur · 4,503 tokens

honcho CLAUDE.md

Claude Code instructions for plastic-labs/honcho, covering claude.md, honcho overview, what is honcho?, core concepts and peer paradigm.

plastic-labs/honcho · 5,828 tokens

Dragon-Brain CLAUDE.md

Instructions for iikarus/Dragon-Brain, covering dragon brain — claude.md, the harness, audit remediation (april–may 2026, complete 2026-05-09), the lie this audit closed and the contract that matters now.

iikarus/Dragon-Brain · 6,132 tokens

brainlayer AGENTS.md

AGENTS.md instructions for EtanHey/brainlayer, covering brainlayer — a letter from etan, brainlayer agent notes (operational), review guidelines, key paths and database.

EtanHey/brainlayer · 4,939 tokens