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/herry2059/project-os-for-codex/agents-mdgit clone --depth 1 https://github.com/herry2059/project-os-for-codexWrote 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/herry2059/project-os-for-codex/agents-md)<a href="https://agentmods.dev/instructions/herry2059/project-os-for-codex/agents-md"><img src="https://agentmods.dev/badge/instructions/herry2059/project-os-for-codex/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 | $0.00656 | $0.00656 |
| Opus 5 | $0.00328 | $0.00328 |
| Sonnet 5 | $0.00131 | $0.00131 |
| Haiku 4.5 | $0.00066 | $0.00066 |
Grade A, and why
project-os-for-codex 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 — 45 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AI contributor rules
These rules apply to Codex and every contributor working in this repository.
Product truth
- This is a real open-source project, not a UI demo. Do not add fake success states, fake production data, or claims that are not implemented.
- Project OS for Codex makes AI-assisted projects visible, resumable, and handoff-ready through project context, Git evidence, next actions, and structured handoffs.
- Work one vertical slice at a time. Finish and verify one end-to-end path before widening the surface.
Architecture and security
- Every persisted business record must belong to a
workspaceId. Requests must derive the workspace from a verified session or AI credential, never from an untrusted body field. - AI credentials must be short-lived, revocable, hashed at rest, scoped to the minimum project and actions, and covered by audit logs.
- Never put passwords, access tokens, provider keys, private endpoints, customer data, or production exports in source, logs, screenshots, prompts, issues, or commits.
- AI may read context and prepare drafts. Payment, deletion, role changes, credential management, publication, deployment, and rollback always require explicit human confirmation.
- Public error messages must not reveal private providers, internal routes, balances, or infrastructure details.
Development workflow
- Read
README.md,docs/INTERFACES.md, and the relevant source before changing behavior. - Preserve user changes and keep commits focused on the requested slice.
- Use
pnpmfor dependency and build commands. - Before committing, run
pnpm run checkand inspect the real affected UI in both light and dark themes. - A change is complete only when the real path works end to end. A diff or a mocked response is not acceptance evidence.
Agent integration contract
- Prefer MCP or documented API/CLI tools over browser automation.
- Never ask users to paste their website password into an AI conversation.
- Project event writes require
Idempotency-Key, server-derived actor identity, input validation, workspace and project isolation, and an audit event. - Keep dangerous tools out of the MCP surface until a human approval workflow exists.
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 · 45 lines · 656 tokens per session scan A 1fdfbb9de211
project-os-for-codex AGENTS.md is an instructions file published in the GitHub repository herry2059/project-os-for-codex (101 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 656 tokens to every session, about $0.0033 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
flow CLAUDE.md
Claude Code instructions for Facets-cloud/flow, covering flow — repo conventions, what this is, build and test, build (produces ./flow in the repo dir, which is on path) and or: go build -o flow .
coco-workflow CLAUDE.md
Instructions for skullninja/coco-workflow, covering claude.md -- coco, project overview, architecture, key files and tracker (lib/tracker.sh).
Atomic-Spec AGENTS.md
Instructions for Chappygo-OS/Atomic-Spec, covering agents.md — adding a new ai agent to atomic spec, about atomic spec and the atomicspec cli, general practices, adding new agent support and current supported agents.
Maestro CLAUDE.md
Instructions for aaryansinha16/Maestro, covering maestro — claude.md, project identity, core operating principles, how it works and the conductor (brain).
product-mode CLAUDE.md
Instructions for sohaibt/product-mode, covering claude.md product-led collaboration guidelines, pre-flight checklist, the seven principles, 1. frame the problem before the solution and 2. make assumptions & unknowns visible.
PM-operating-OS AGENTS.md
Instructions for hardiktiwari/PM-operating-OS, covering pm chief of staff, how to think, knowledge base, onboarding and communication style.