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/muratgur/ordinus/agents-mdgit clone --depth 1 https://github.com/muratgur/ordinusWhat 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 | $0.01112 | $0.01112 |
| Opus 5 | $0.00556 | $0.00556 |
| Sonnet 5 | $0.00222 | $0.00222 |
| Haiku 4.5 | $0.00111 | $0.00111 |
Grade A, and why
ordinus 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 — 114 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Ordinus Agent Guide
Mission
Ordinus is a desktop workspace for coordinating AI agents around real software work. Keep the product simple, observable, local-first, and user-controlled.
Build a command center for AI-assisted work, not a generic chatbot.
Product Principles
- Make agent activity visible: planned, running, blocked, completed.
- Give agents clear roles and responsibilities.
- Help agents coordinate instead of running disconnected conversations.
- Keep the user in control of important actions.
- Prefer a small reliable workflow over broad unfinished features.
- Avoid integrations before the core runtime model is stable.
Architecture Direction
- Electron main process owns privileged work: filesystem, database, process execution, provider runtimes, and OS integration.
- Renderer is UI only. It must not access Node, filesystem, child processes, secrets, or SQLite directly.
- Preload exposes only a small typed
window.ordinusAPI. - Never expose raw
ipcRenderer,electron, filesystem, process, or database APIs to renderer code. - Provider runtime must follow
docs/provider-runtime-contract.mdbefore adding CLI detection or process spawning. - Provider commands use executable plus args, not raw shell strings.
- Runtime environment is deny-by-default; secrets are referenced, resolved in main, and never logged.
- Shared contracts live in
src/shared. - Use SQLite for durable app state, but add tables only when the product model is clear.
- Keep SQLite schema changes in Drizzle migrations under
app/resources/db/migrations. - Run migrations from the Electron main process before registering IPC handlers.
- Track applied SQL through Drizzle and app-level schema version through
app_meta.schema_version. - Do not add agent, task, provider, schedule, or inbox schemas prematurely.
Cross-Platform Development
Ordinus must work on both macOS and Windows. When making changes, account for platform differences before finishing.
- Do not assume POSIX-only paths, shell commands, environment variable syntax, executable names, file permissions, or case-sensitive filesystems.
- Use Node/Electron APIs such as
path,os, and provider commandexecutable + argsarrays instead of shell-specific strings. - Consider Windows
.cmd/.exebehavior, spaces in paths, CRLF/LF differences, and macOS/Linux case sensitivity. - For changes touching filesystem access, process execution, packaging, provider runtimes, scripts, or native dependencies, verify or explain platform impact.
- Prefer npm scripts and cross-platform tooling over shell-specific commands.
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 · 114 lines · 1,112 tokens per session scan A 386f05d22497
ordinus AGENTS.md is an instructions file published in the GitHub repository muratgur/ordinus (110 stars, last pushed 21d ago), licensed MIT. It adds 1,112 tokens to every session, about $0.0056 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
AgentCore AGENTS.md
Instructions for Lawofall/AgentCore, covering agents.md, 先读哪里, docs 与 rules and 开发 / 测试(最短).
wayland AGENTS.md
AGENTS.md instructions for FerroxLabs/wayland, covering coordination (read every task — multi-agent blackboard), wayland - project guide, code conventions, file & directory structure and naming.
wayland copilot-instructions.md
Copilot instructions for FerroxLabs/wayland, covering ijfw rules, output discipline, memory routing, context discipline and cross-audit.
wayland CLAUDE.md
Claude Code instructions for FerroxLabs/wayland, a project described as: Wayland - The AI Agent That Perceives. Reasons. Acts. Evolves.
agent-desktop CLAUDE.md
Claude Code instructions for BaLaurent/agent-desktop, covering agent desktop — project instructions, build & run, architecture decisions, conventions & cascade and ordering constraints.
native-cli-ai AGENTS.md
AGENTS.md instructions for madebyaris/native-cli-ai, covering learned user preferences and learned workspace facts.