swarm-legacy: Instructions file for Claude Code

CLAUDE.md

swarm-legacy CLAUDE.md is an instructions file for Claude Code from miopea/swarm-legacy. It costs 2,809 tokens per session, scanned B, original, MIT.

Project instructions for Swarm (legacy), a Python program that coordinates multiple Claude Code agents through a background service and terminal workers. The project is maintained for compatibility and has been replaced by Swarm Next.

In plain words
What is it for?
Use them when maintaining Swarm (legacy), running its commands or service, relocating an installation, or investigating its documented database traps and project layout.
Why use it?
They explain the project's unusual legacy naming, installation locations, service name, and relocation behavior. This prevents maintenance work from being confused with work on the newer project.

Instructions file for Claude Code

Written for Claude Code: the file is CLAUDE.md. Also seen: reads .claude/ paths; mentions CLAUDE.md; mentions Claude Code.

This is miopea/swarm-legacy's own configuration. It tells Claude Code how to work on swarm-legacy itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything swarm-legacy configures →

Reuse

Borrowing it

Nothing to install: this file belongs to miopea/swarm-legacy. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/miopea/swarm-legacy/main/CLAUDE.md
Clone the repo
git clone --depth 1 https://github.com/miopea/swarm-legacy

Made for: Claude Code.

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 swarm-legacy CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/miopea/swarm-legacy/claude-md.svg)](https://agentmods.dev/instructions/miopea/swarm-legacy/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/miopea/swarm-legacy/claude-md"><img src="https://agentmods.dev/badge/instructions/miopea/swarm-legacy/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 2,809 This file is loaded in full into every session.
When invoked 2,809 The same file — it is already loaded in full.
Security scan B 2 findings. A grade says what 26 rules found in the file — not that it is safe.
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 $0.02809 $0.02809
Opus 5 $0.01404 $0.01404
Sonnet 5 $0.00562 $0.00562
Haiku 4.5 $0.00281 $0.00281

Measured 8d ago against content hash 8043eac3576b, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade B, and why

swarm-legacy CLAUDE.md scanned grade B with 2 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 8d 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.

Asks for rootmediumPrivilege escalation

A mod that escalates privileges can change anything on the machine, not only the project.

`systemctl --user`, not `sudo systemctl`.

Makes network callslowCapability

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

Need to verify? → Actually run the query/curl/command. Never assume.
CLAUDE.md · 206 lines

How it starts

The opening of the file, as written. The whole thing — 206 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Swarm (legacy) — Project Guide

See ~/.claude/CLAUDE.md for universal rules (code quality, verification, shipping vocabulary, secrets, swarm) and its routing map to rcg-architecture/docs/standards/.

Orientation

  • What this is. swarm-aiSwarm (legacy), a hive-mind orchestrator for Claude Code agents. Python ≥3.12, aiohttp daemon plus PTY workers, and two console scripts (swarm and swarm-legacy) that are the same entry point. Superseded by Swarm Next; this repo is maintenance-only. The GitHub repo is miopea/swarm-legacy, and this checkout now lives at ~/projects/personal/swarm-legacy~/projects/personal/swarm was freed for Swarm Next on 2026-08-21. Display strings say "Swarm (legacy)"; the package name does not — it is still swarm-ai on PyPI. The installed names do move, but only when the operator runs swarm relocate: swarmswarm-legacy, swarm.serviceswarm-legacy.service, ~/.swarm~/.swarm-legacy. Both entrypoints ship together so an un-relocated install keeps working.
  • How it runs here. This box is relocated and now a production install: systemd user unit swarm-legacy.service, ExecStart=~/.local/bin/swarm-legacy serve, WorkingDirectory=$HOME, daemon API on :9090, state in ~/.swarm-legacy. It no longer runs from this checkout — the tool is installed from git, so the source tree is for editing only and uv run is not in the startup path. Not a system unit — systemctl --user, not sudo systemctl.
  • swarm-next is a DIFFERENT system on the same box. swarm-next-api.service and swarm-next-terminal-host.service run alongside this one from ~/projects/personal/swarm-next. Restarting the wrong unit looks like a fix that changed nothing.
  • All state is one SQLite file, in the state dir~/.swarm-legacy/swarm.db after swarm relocate, ~/.swarm/swarm.db before it (~99 MB). Never hardcode either: call swarm.paths.state_dir(), which prefers the relocated directory. Beside the DB, invariant-log.jsonl (Tier 0 hook decisions) and pty-writes.jsonl. Nothing authoritative lives in this repo's tree.
  • No deploy. There is no production target and no deploy.md. Push is the terminal state — say pushed, never shipped.
  • CI gates on two Python versions, test (3.12) and test (3.13), and they are required checks on main. Both must be green before merge.

Read the full file on GitHub · 206 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. 8d ago First seen · 206 lines · 2,809 tokens per session scan B 8043eac3576b

Subscribe to this mod's changes

swarm-legacy CLAUDE.md is an instructions file published in the GitHub repository miopea/swarm-legacy (9 stars, last pushed 9d ago), licensed MIT. It adds 2,809 tokens to every session, about $0.0140 per session on Opus 5. A static security scan graded it B with 2 findings (asks for root, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other instructions, from other repositories

next.js AGENTS.md

AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.

vercel/next.js · 7,296 tokens

codex AGENTS.md

AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.

openai/codex · 5,153 tokens

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 6,785 tokens

vscode oss-third-party-notices.instructions.md

Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).

microsoft/vscode · 5,001 tokens

langchain AGENTS.md

AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.

langchain-ai/langchain · 4,469 tokens

spec-kit AGENTS.md

AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.

github/spec-kit · 7,104 tokens