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/agno-agi/agentos-render/agents-mdgit clone --depth 1 https://github.com/agno-agi/agentos-renderWhat 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 | $0.15295 | $0.15295 |
| Opus 5 | $0.07648 | $0.07648 |
| Sonnet 5 | $0.03059 | $0.03059 |
| Haiku 4.5 | $0.01529 | $0.01529 |
Grade C, and why
agentos-render AGENTS.md scanned grade C 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.
Harvests environment variableshighData exfiltration
Enumerating or grepping the environment for keys collects credentials unrelated to what the mod says it does.
**What it will not do.** It refuses unsafe capability before calling a tool — secret exfiltration, reading `.env`, unrestricted writes, shell — and it screens the instructions it writes, not only the tools it wires, beca This is a copy
89% identical to agentos-docker AGENTS.md — 56 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 — 374 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AgentOS — Render template
This file is the source of truth for any agent (Claude Code, Codex, others) working in this repo. CLAUDE.md is a symlink to this file — edit one, both update.
Project Overview
AgentOS: the agent platform that powers your product. AgentOS is an agent server built on the Agno framework that turns your agents into a production API attaching to any client: REST API for programmatic use, chat interfaces for humans (Slack is wired in; WhatsApp/Telegram/Discord mirror the same pattern), and MCP at /mcp for AI apps (claude.ai, ChatGPT, Cursor, Claude Code) — which work through the platform, not just on it. The platform grows through two lanes. Lane 1: coding agents change the source — eight coding-agent skills cover platform setup, the full agent development lifecycle, and the production deploy, governed by git. Lane 2: the platform builds at runtime — Platform Builder composes agents, teams, and workflows from a safe registry of reviewed blocks, governed by the Studio catalog (drafts, publish, versions, archive/restore). app/registry.py is the membrane between the lanes: every capability a built component can carry is declared there by a reviewed code change, and the builder composes but never expands it. Three platform agents run the show — Platform Builder (builds at runtime), Platform Manager (watches the runtime), Platform Engineer (reads the source) — and Agno, the team that fronts them: the platform speaking for itself, the one name everybody tags in ("Agno, what's happening with radar?" — "Agno, build me an agent for this"). It holds the thread — people, projects, decisions, living notes — learns how each user works, and answers with the state of play from Slack, claude.ai, ChatGPT, or any MCP client, with everything built at runtime one runner call away. Postgres (pgvector) handles persistence for sessions, memory, and knowledge. Runs locally via Docker; this template deploys to Render via the Blueprint in render.yaml plus a wiring script and is the Render sibling of the agentos-* deployment family — see Portable core vs. deploy layer.
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 · 374 lines · 15,295 tokens per session scan C 8af7a81e6c45
agentos-render AGENTS.md is an instructions file published in the GitHub repository agno-agi/agentos-render (2 stars, last pushed 2d ago), licensed Apache-2.0. It adds 15,295 tokens to every session, about $0.0765 per session on Opus 5. A static security scan graded it C with 1 finding (harvests environment variables). It is 89% identical to agentos-docker AGENTS.md, differing in 56 lines, and is treated as a copy.
Other instructions, from other repositories
agentos-docker AGENTS.md
Instructions for agno-agi/agentos-docker, covering agentos — docker template, project overview, architecture, key files and development setup.
agno-agent-starter AGENTS.md
Instructions for linhai0872/agno-agent-starter, covering agno agent service 开发规范, 核心哲学, 项目结构概览, 开发规范 and 命名规则.
agno-agent-starter CLAUDE.md
Instructions for linhai0872/agno-agent-starter, covering claude code 开发规范, 引用主规范, claude 专用说明, 文件编辑 and 命令执行.
agentos-docker CLAUDE.md
Instructions for agno-agi/agentos-docker, a project described as: AgentOS: Serve agents over API, MCP, and chat interfaces like Slack.
agentos-modal CLAUDE.md
Instructions for agno-agi/agentos-modal, a project described as: AgentOS: Serve agents over API, MCP, and chat interfaces like Slack.
agentos-modal AGENTS.md
Instructions for agno-agi/agentos-modal, covering agentos — modal template, project overview, architecture, key files and development setup.