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/tools-for-agents/agent-hq/agents-mdgit clone --depth 1 https://github.com/tools-for-agents/agent-hqWrote 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.
[](https://agentmods.dev/instructions/tools-for-agents/agent-hq/agents-md)<a href="https://agentmods.dev/instructions/tools-for-agents/agent-hq/agents-md"><img src="https://agentmods.dev/badge/instructions/tools-for-agents/agent-hq/agents-md.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.01041 | $0.01041 |
| Opus 5 | $0.00521 | $0.00521 |
| Sonnet 5 | $0.00208 | $0.00208 |
| Haiku 4.5 | $0.00104 | $0.00104 |
Grade A, and why
agent-hq 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 3d 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 — 75 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — agent-hq
🛰️ The operating platform for an all-agent company. Shared memory, a kanban agents claim work from (atomic claim/lease so parallel agents don't collide), an agent registry, messaging, a run/cost ledger, a live dashboard with a knowledge-graph tab, and an MCP server exposing all of it (28 tools). Part of tools-for-agents.
Setup
node --version # 22+ required. Nothing to install.
npm test # = node --test (~12s)
HQ_DB_PATH=/tmp/hq/hq.db PORT=9310 node src/server.js &
HQ_URL=http://localhost:9310 npm run seed # 70 tasks, 20 agents, a real graph
npm run mcp # the MCP server, stdio
Zero runtime dependencies, and that is a hard rule. No dependencies in package.json, ever. Node 22+
gives you node:sqlite and a test runner.
| Env | For |
|---|---|
HQ_DB_PATH |
the SQLite file (default ./data/agenthq.db) — always redirect this in tests |
PORT |
server port (default 7700) |
HQ_STALE_MS |
how long before an agent's heartbeat is considered stale |
⚠️ :7700 is usually a live instance someone is actually using. Use 9310+ for anything you are testing,
with your own HQ_DB_PATH. Never point a test at a running deployment.
⚠️ pkilling the server leaves data in an uncheckpointed -wal — the next server on the same path reads
zero rows and looks like a corruption bug. Just re-seed.
The rules this repo is built on
1. Only the picture is evidence. Run iris against any UI change and look at the shot. This dashboard's gate audited a blank board for months and went green every time — the seed script existed; nobody ran it. Seed, assert the data is on the page, then look.
2. Open the doors. Half this app's hover surface lives behind buttons no gate presses: modals, the task
detail, the graph panel. Everything found behind one so far was a bug. The graph panel opens by clicking a
node on a canvas — window.HQGraph.debug() returns page-space node centres for exactly this.
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.
- 3d ago First seen · 75 lines · 1,041 tokens per session scan A 4353225a8381
agent-hq AGENTS.md is an instructions file published in the GitHub repository tools-for-agents/agent-hq (0 stars, last pushed 29d ago), licensed MIT. It adds 1,041 tokens to every session, about $0.0052 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 2026-08-31.
Other instructions, from other repositories
kaneo CLAUDE.md
Claude Code instructions for usekaneo/kaneo, a project described as: 🎯 All you need. Nothing you don't. Open source project management that works for you, not against you.
veritas-kanban AGENTS.md
AGENTS.md instructions for BradGroux/veritas-kanban, covering agents.md — canonical agent instructions for veritas kanban, runtime requirements, repository layout, essential commands and install.
kanban CLAUDE.md
Claude Code instructions for kanban-rs/kanban, covering claude.md, project overview, architecture philosophy, solid principles applied and workspace structure.
flux AGENTS.md
Instructions for sirsjg/flux, covering agents.md, project overview, dogfooding, common commands and development.
claude-session-dashboard CLAUDE.md
Claude Code instructions for dlupiak/claude-session-dashboard, covering claude.md, mandatory: agent delegation, delegation rules (never skip these), what you (the main context) may do directly and what you must never do directly.
oh-my-symphony AGENTS.md
Instructions for cskwork/oh-my-symphony, covering agents.md: codex cli entry point, source of truth: skills/symphony-skill, available skill (operator-facing), symphony-skill and worker-side guidance.