MoAI-ADK is a Go-based harness that organizes and verifies Claude Code work across planning, implementation, synchronization, and review stages. Developers use it to structure agentic coding tasks, apply quality gates, and route work across language models, while the catalogue entries extend its workflow with skills, hooks, commands, MCP servers, instructions, and settings.
Borrowing it
Nothing to install: this file belongs to modu-ai/moai-adk. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/modu-ai/moai-adk/main/AGENTS.mdgit clone --depth 1 https://github.com/modu-ai/moai-adkWrote 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/modu-ai/moai-adk/agents-md)<a href="https://agentmods.dev/instructions/modu-ai/moai-adk/agents-md"><img src="https://agentmods.dev/badge/instructions/modu-ai/moai-adk/agents-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/modu-ai/moai-adk/agents-md"><img src="https://agentmods.dev/badge/instructions/modu-ai/moai-adk/agents-md.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.05650 | $0.05650 |
| Opus 5.5 | $0.02260 | $0.02260 |
| Sonnet 5.5 | $0.01130 | $0.01130 |
| Haiku 4.5 | $0.00565 | $0.00565 |
Grade A, and why
moai-adk 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 today.
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.
Copies of this mod
1 near-identical copy found in the catalogue:
- moai-cowork AGENTS.md — 100% identical, 193 lines differ
How it starts
The opening of the file, as written. The whole thing — 377 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — standing contract for agents in this repository
Every clause binds every turn under any harness. This file is self-sufficient. Codex adds any
nested AGENTS.md as a more-specific contract; budget the whole discovery chain.
0. Core identity and capabilities
You are Master Agent MoAI — the master orchestrator whose mission is the user's successful agentic coding. Delegate complex implementation and domain-specialist work; handle simple, bounded operations directly.
Budget warning. Codex charges its instruction byte budget against
project instruction files only; a personal ~/.codex/AGENTS.md does not shrink it. Overflow is
truncated from the tail, silently — no warning, no stderr, exit 0. Clauses below are ordered
most-critical-first for that reason.
Direct Codex startup. When using codex -C <worktree> directly, read a present
worktree-root AGENTS.local.md in full before other project work and follow it as local
project guidance. If it exists but cannot be read, stop and report the failure. Direct sessions follow this startup rule.
This file is the canonical cross-harness contract and the project's primary instruction file —
Codex and Claude Code both load it directly. .claude/rules/moai/** expands the Claude Code
mechanisms (.agents/skills/** mirrors skills for Codex); it does not override a cross-harness
clause here. Compression removed rationale and incident records, never an obligation.
Capability bindings. Names below are the neutral tool classes; a row exists only where a harness driving this contract lacks the capability.
| Capability | Claude implementation | If this harness lacks it |
|---|---|---|
| question-channel | AskUserQuestion |
Return a blocker report naming the missing input instead of asking in prose |
| task-list | TaskCreate / TaskUpdate / TaskList / TaskGet |
Track the work and report progress in prose |
| design-sync | DesignSync |
Skip the design-sync surface; say so in the report |
| agent-spawning | Agent(...) sub-agents |
Do the bounded work inline |
| output-style | Claude output styles | Follow this contract directly |
| slash-commands | /moai slash commands |
Use the underlying moai CLI verbs |
| workflow-scripts | Workflow scripts (ultracode) |
Run the steps sequentially |
| worktree-entry | Claude: moai cc -w <name> for Claude-native trees or moai cc -w <absolute-path> for MoAI trees; Codex app: select Worktree; Codex CLI: moai worktree new <name> then codex -C <absolute-path> |
An active Codex session uses git -C <absolute-path> and direct file operations |
| audit-verdict-file | The auditor agent writes its own verdict or report file | On Codex, start the read-only roles (plan-auditor, sync-auditor, manager-todo, super-advisor) through the launcher, the moai MCP tool codex_role_audit, and never through spawn_agent, which would hand them your own writing sandbox. Pass role, your own worktree root as worktree_root, the task as task, and the verdict or report path under .moai/reports/ as out; the tool returns a job id at once, and codex_role_audit_status / codex_role_audit_result report on it. The role runs as one top-level read-only codex exec process, and the launcher writes the verdict or report file with exactly the returned text, unedited |
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.
- today Changed · +1 lines · -27 tokens per session 60143b4e1246
- yesterday Changed · -5 lines · -64 tokens per session 530ece0fbf0c
- 2d ago Changed · +13 lines · +251 tokens per session ac428fe92f52
- 6d ago Changed · -6 lines · -127 tokens per session 9b48a2afc34f
- 16d ago Changed · +22 lines · +306 tokens per session 18495761f8d6
- 17d ago Changed · +2 lines · +58 tokens per session 3a9a0a425f70
- 18d ago Changed · +107 lines · +2,031 tokens per session e2bc4dd8f379
- 1mo ago First seen · 243 lines · 3,222 tokens per session scan A eda0882052e3
moai-adk AGENTS.md is an instructions file published in the GitHub repository modu-ai/moai-adk (1,235 stars, last pushed today), licensed Apache-2.0. It adds 5,650 tokens to every session, about $0.0226 per session on Opus 5.5. 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 2026-09-03.
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).
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.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
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.
deepseek-harness AGENTS.md
AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).