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/smaramwbc/statewave/agents-mdgit clone --depth 1 https://github.com/smaramwbc/statewaveWhat 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 | $0.00933 | $0.00933 |
| Opus 5 | $0.00466 | $0.00466 |
| Sonnet 5 | $0.00187 | $0.00187 |
| Haiku 4.5 | $0.00093 | $0.00093 |
Grade A, and why
statewave AGENTS.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 2d 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 — 81 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — guide for contributors and coding agents
A short orientation for humans and AI coding agents (GitHub Copilot, Claude, Cursor, …) working in the statewave core repo. For the full contribution process and licensing, start with CONTRIBUTING.md.
What this repo is
Statewave is an open-source memory/context service for AI agents: it records
episodes, compiles them into durable memories, and serves compact, ranked,
token-bounded context over a /v1 API. This repo is the server and reference
implementation. For the big picture, see the
architecture overview.
Setup, build, test
The canonical steps live in CONTRIBUTING.md and the README Quickstart — don't duplicate them, follow them. In short:
docker compose up db -d
python -m venv .venv && source .venv/bin/activate
pip install -e ".[dev,llm]"
alembic upgrade head
pytest tests/
Run ruff and the test suite before opening a PR; make test-cold validates a
full fresh-install path.
Tip for agents: unit tests are hermetic with stub providers — run them without external services via:
STATEWAVE_EMBEDDING_PROVIDER=stub STATEWAVE_COMPILER_TYPE=heuristic \
pytest tests/ --ignore=tests/integration
Conventions
- Code style & testing: see statewave-docs/dev/conventions.md.
- Packages version independently. The server, the Python SDK (
statewave), and the TypeScript SDK (@statewavedev/sdk) version on separate cadences; the compatibility axis is the/v1API contract, not a shared number. Don't try to align version strings across packages. - Some "proof" numbers are mirrored across repos — test counts, eval
assertion/test counts, and the support-workflow benchmark score. They have a
single source of truth and a consistency check in
statewave-docsthat runs at release time. If you change one of these numbers, keep every surface in sync rather than editing a single file. - Keep claims accurate and modest. Describe what the code does; back any performance or benchmark claim with a reproducible source, and avoid unqualified superlatives.
- CI enforces a few cross-cutting invariants — tenant-scoped repository queries, bounded pagination params, and a single shared tokenizer. If a guard test fails, satisfy the rule rather than weakening the test. See CONTRIBUTING.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.
- 2d ago First seen · 81 lines · 933 tokens per session scan A e51d7e751aa2
statewave AGENTS.md is an instructions file published in the GitHub repository smaramwbc/statewave (318 stars, last pushed 2d ago), licensed Apache-2.0. It adds 933 tokens to every session, about $0.0047 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
LaunchStack CLAUDE.md
Instructions for Deodat-Lawson/LaunchStack, covering claude.md, frontend rules (apps/web, apps/landing) and workflow.
interview-guide AGENTS.md
AGENTS.md instructions for Snailclimb/interview-guide, covering ai interview platform agent rules, tech stack, commands, project structure and architecture.
interview-guide CLAUDE.md
Claude Code instructions for Snailclimb/interview-guide, covering claude code instructions, path rules and maintenance.
opencrawling CLAUDE.md
Instructions for opencrawling/opencrawling, covering claude.md, project overview, build commands, full build (all maven modules) and run the embedding microservice (separate process).
vector-mcp AGENTS.md
Instructions for Knuckles-Team/vector-mcp, covering agents.md, scope, current architecture, non-negotiable design rules and source layout.
pgmnemo AGENTS.md
Instructions for pgmnemo/pgmnemo, covering pgmnemo — agent integration guide, 1. what pgmnemo is — and the problem it solves, the problem, architectural consequences and 2. when to adopt / when not to.