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/w48669664-bit/colla-agent/agents-mdgit clone --depth 1 https://github.com/w48669664-bit/colla-agentWrote 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/w48669664-bit/colla-agent/agents-md)<a href="https://agentmods.dev/instructions/w48669664-bit/colla-agent/agents-md"><img src="https://agentmods.dev/badge/instructions/w48669664-bit/colla-agent/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 | $0.00425 | $0.00425 |
| Opus 5 | $0.00212 | $0.00212 |
| Sonnet 5 | $0.00085 | $0.00085 |
| Haiku 4.5 | $0.00042 | $0.00042 |
Grade A, and why
colla-agent 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 4d 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.
What it actually says
Relay Room
This repository provides a transparent collaboration control plane for a current Codex session and the Antigravity CLI. Snake remains the reference workload used by repository smoke tests; it is not the product boundary.
Roles
- In Skill-driven host mode, the current Codex session is the only
implementation agent. Never invoke
codex,codex exec, another Codex session, a subagent, or any recursive agent workflow. - Antigravity is always a read-only planner and reviewer. It must not edit project files or invoke other agents.
- Record meaningful prompts, explicit responses, command summaries, file changes, tests, and phase transitions. Never record credentials or claim to expose private chain-of-thought.
Iteration policy
- Supported review caps are 1, 3, 5, 8, and 12 rounds. Default to 5.
- Let Antigravity recommend a budget within the selected cap.
- Stop early on PASS.
- Stop on three consecutive repeated must-fix signatures, the selected cap, a manual stop, or a blocking failure.
- Never run an infinite review loop or silently raise the cap.
Generic deliverables
- Keep writes inside the active user-selected project.
- Register meaningful outputs in Relay Room after implementation.
- Preview HTML directories interactively; use native previews for images, video, audio, PDF, and text where possible.
- Persist each host run under
.relay-room/runs/<run-id>/in the active project.
Reference Snake deliverable
- Keep the standalone Canvas game in
public/game/. - Preserve keyboard and touch controls, score, best score, pause/resume,
restart, game-over state, and deterministic
window.__snakeDebug. - Keep the reference game dependency-free.
Repository validation
- Run
npm run lint. - Run
npm test. - Run
node tests/smoke-game.mjs. - Validate
skills/relay-room-collaboration/with the Skill validator.
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.
- 4d ago First seen · 49 lines · 425 tokens per session scan A d1ffeebe53bd
colla-agent AGENTS.md is an instructions file published in the GitHub repository w48669664-bit/colla-agent (2 stars, last pushed 2d ago), licensed MIT. It adds 425 tokens to every session, about $0.0021 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
openhuman AGENTS.md
AGENTS.md instructions for tinyhumansai/openhuman, covering openhuman, repository layout, runtime scope, ios client (experimental, non-shipping) and commands (from repo root).
Lody AGENTS.md
AGENTS.md instructions for LodyAI/Lody, covering repository guidelines, context maintenance, repository boundary, project map and checks and commits.
openhuman CLAUDE.md
Claude Code instructions for tinyhumansai/openhuman, a project described as: Your Personal AI super intelligence. A brain that builds a local-first memory of your life, a fantastic orchestrator of agent fleets and workflows, and a deep researcher.
open-tag CLAUDE.md
Instructions for fancyboi999/open-tag, a project described as: Open-source, self-hostable alternative to Claude Tag — a Slack-style workspace where your team and its AI agents (Claude Code, Codex, GitHub Copilot, and more) work as teammates in channels, threads, DMs, and shared tasks. Your data stays on your machines.
rondoflow CLAUDE.md
Claude Code instructions for rondoflow/rondoflow, covering rondoflow, project overview, tech stack, development and setup.
open-tag AGENTS.md
Instructions for fancyboi999/open-tag, covering open-tag — guide for ai coding agents, this file is a map, not a manual, conventions, parallel development (worktrees) and isolated dev e2e (on demand).