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/tu11aa/squadrant/gemini-mdgit clone --depth 1 https://github.com/tu11aa/squadrantWhat 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 | $0.02414 | $0.02414 |
| Opus 5 | $0.01207 | $0.01207 |
| Sonnet 5 | $0.00483 | $0.00483 |
| Haiku 4.5 | $0.00241 | $0.00241 |
Grade B, and why
squadrant GEMINI.md scanned grade B 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 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.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
## Managed `~/.claude/settings.json` (#615) This is a copy
94% identical to squadrant AGENTS.md — 9 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 92 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Project Direction: Multi-Agent
Squadrant is a multi-agent orchestration layer, not a Claude-Code-only tool. Claude Code is the reference implementation today; Codex, Cursor, and Gemini CLI are supported (or in progress) through the runtime driver abstraction and the upcoming cross-agent projection layer (issue #31).
When working on squadrant:
- Prefer
AGENTS.mdas the canonical instruction format.CLAUDE.mdis becoming a thin wrapper. - When adding agent-facing features, ask: "does this work for non-Claude agents too?" If not, file a follow-up issue to generalize it.
- Don't add Claude-only surface area without a migration path. The three plugin slots (runtime / workspace / notifier) exist specifically to avoid this.
- Skills in
plugin/skills/are portable markdown — Claude Code reads them via the Skill tool; other agents read them viaAGENTS.mdinclusion.
Full direction statement: docs/specs/2026-04-24-multi-agent-direction.md.
Repository layout
Six packages in a one-way DAG: shared ◄ core ◄ {agents, workspaces, web} ◄ cli
| Package | Owns |
|---|---|
@squadrant/shared |
Config schema, types, constants — leaf, zero internal deps |
@squadrant/core |
Daemon, state-machine, protocol, AgentDriver interface |
@squadrant/agents |
AI driver seam: claude / codex / opencode / gemini |
@squadrant/workspaces |
Runtime (cmux), workspace (obsidian), notifier drivers |
@squadrant/web |
Observability dashboard (bundled HTML/JS) |
@squadrant/cli |
Commands, bin entry, daemon host — root package |
Build outputs: dist/index.js (CLI bin) · dist/squadrantd.js (daemon). See architecture diagram.
Telegram (opt-in, #65)
Two-way Telegram lives in @squadrant/core (src/telegram/*: client/format/state/bridge/setup) and is wired into the daemon by the CLI host (squadrantd.ts) — a daemon-internal TelegramBridge, not a separate process. Outbound crew lifecycle events push to a per-project forum topic; inbound replies become a captain.message mailbox entry delivered to the captain pane. It is constructed only when config.telegram is present (zero behavior change otherwise) and uses plain fetch — no runtime SDK (@grammyjs/types is a dev-only type dep). Set up via squadrant telegram setup (interactive wizard) then squadrant telegram link <project> / squadrant telegram status; full guide + config block in docs/reference.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 · 92 lines · 2,414 tokens per session scan B 4b281e838b43
squadrant GEMINI.md is an instructions file published in the GitHub repository tu11aa/squadrant (2 stars, last pushed 2d ago), licensed MIT. It adds 2,414 tokens to every session, about $0.0121 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). It is 94% identical to squadrant AGENTS.md, differing in 9 lines, and is treated as a copy.
Other instructions, from other repositories
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).
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.
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.
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.
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).
next.js AGENTS.md
Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.