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/xvirobotics/metabot/agents-mdgit clone --depth 1 https://github.com/xvirobotics/metabotWrote 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/xvirobotics/metabot/agents-md)<a href="https://agentmods.dev/instructions/xvirobotics/metabot/agents-md"><img src="https://agentmods.dev/badge/instructions/xvirobotics/metabot/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 | $0.00649 | $0.00649 |
| Opus 5 | $0.00324 | $0.00324 |
| Sonnet 5 | $0.00130 | $0.00130 |
| Haiku 4.5 | $0.00065 | $0.00065 |
Grade A, and why
metabot 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 5d 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 — 70 lines — stays where its author put it; the contents beside it link to each section on GitHub.
MetaBot Personal Edition Workspace
MetaBot connects chat-based agents with self-hosted Memory, Skills, Agent Bus,
Agent Teams, T5T, scheduling, and runtime operations. Use metabot as the
single CLI for these surfaces; legacy mm, mh, and mb workflows are
retired.
Core CLI
metabot memory search "query"
metabot memory get <id|path>
metabot skills list
metabot agents list
metabot agents talk <agent> "message"
metabot inbox poll --loop
metabot teams status <team> --summary
metabot t5t push <project> <date> "entry"
Use the metabot Skill for command details and the metabot-team Skill for
durable Team state. Personal Core and its web console default to
http://127.0.0.1:9200; users may configure another Core with
METABOT_CORE_URL.
AGENTS.md is the canonical project instruction file. Existing workspace
AGENTS.md and CLAUDE.md files are user-owned and MetaBot install/update must
not replace them.
Working Principles
- Start from the user's goal and inspect the actual workspace or runtime before acting.
- Carry authorized work through implementation and verification appropriate to its risk. Do not stop at a plan unless the user asked only for a plan.
- Keep changes focused and preserve unrelated files, local modifications, user data, configuration, and user-selected integrations.
- Choose the execution approach that best fits the task. Add coordination or extra process only when it materially helps the outcome.
- Use the injected Current MetaBot Context and Team Context as the current
chat, schedule, roster, and routing source. Query
metabot teamswhen more durable state is needed. - Ask the user only when a missing decision, authority, credential, or unsafe action genuinely blocks progress.
- Never expose secrets or authorization tokens. Avoid destructive actions unless they are explicitly authorized and necessary.
- Before runtime or deployment work, verify the real target, current state, and rollback path. Do not assume host-specific paths, process names, or ports.
- Search before creating shared resources, and keep user-facing updates concise and evidence-backed.
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.
- 5d ago First seen · 70 lines · 649 tokens per session scan A f55e5e53b17a
metabot AGENTS.md is an instructions file published in the GitHub repository xvirobotics/metabot (977 stars, last pushed 3d ago), licensed MIT. It adds 649 tokens to every session, about $0.0032 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 2026-08-30.
Other instructions, from other repositories
openlark AGENTS.md
AGENTS.md instructions for foxzool/openlark, covering openlark knowledge base, structure, where to look, conventions and 模块组织.
feishu-cursor-claw AGENTS.md
Instructions for nongjun/feishu-cursor-claw, covering agents.md — feishu-cursor-claw（虾群平台版）, 项目定位, 技术栈, 目录结构 and 模块职责与依赖.
openlark CLAUDE.md
Claude Code instructions for foxzool/openlark, a project described as: 飞书开放平台的非官方 Rust SDK，支持自定义机器人、长连接机器人、云文档、飞书卡片、消息、群组等 API 调用。.
botmux CLAUDE.md
Claude Code instructions for deepcoldy/botmux, covering botmux, 构建 & 运行, bun 开发链路, worktree 的 nodemodules：共享还是独立（改前必读） and 编译态（单文件二进制）注意.
lodestar AGENTS.md
AGENTS.md instructions for leviyuan/lodestar, covering lodestar 项目指引, 目录, 开发规则, 模块边界 and 运行中的 daemon.
forge-hub CLAUDE.md
Instructions for LinekForge/forge-hub, covering claude.md, 项目概览, monorepo 结构, 构建与测试命令 and hub-server：类型检查 + 单元测试.