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/rixzkiye/codex-router/agents-mdgit clone --depth 1 https://github.com/rixzkiye/codex-routerWrote 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/rixzkiye/codex-router/agents-md)<a href="https://agentmods.dev/instructions/rixzkiye/codex-router/agents-md"><img src="https://agentmods.dev/badge/instructions/rixzkiye/codex-router/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 | $0.00221 | $0.00221 |
| Opus 5 | $0.00111 | $0.00111 |
| Sonnet 5 | $0.00044 | $0.00044 |
| Haiku 4.5 | $0.00022 | $0.00022 |
Grade A, and why
codex-router 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 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.
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.
What it actually says
Codex Router repository contract
Product invariants
- Keep MCP lifecycle calls bounded; a start call must never wait for turn completion.
- Append normalized runtime events before updating projections.
- Never replay a turn after semantic output, a command, a file mutation, an approval, or an external side effect is observed.
- Keep logical-agent, incarnation, runtime, thread, turn, and worktree identity separate.
- A write worktree has one persisted lease owner and a monotonically increasing fencing token.
- Runtime profiles contain secret references only. Never persist resolved credential or
CODEX_HOMEvalues. - Continuations retain runtime and thread affinity. Cross-runtime recovery is an explicit handoff with a new thread.
- Results label worker reports separately from router-observed evidence.
Verification
Run pnpm verify for the full local gate. Use focused pnpm vitest run <path> while iterating.
Protocol compatibility is pinned in src/runtime/codex-protocol.ts. Validate changes against generated types from the supported codex app-server before changing its request shapes.
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 · 19 lines · 221 tokens per session scan A dce9ba86f00b
codex-router AGENTS.md is an instructions file published in the GitHub repository rixzkiye/codex-router (0 stars, last pushed 7d ago), licensed MIT. It adds 221 tokens to every session, about $0.0011 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 2026-08-31.
Other instructions, from other repositories
mission-control CLAUDE.md
Claude Code instructions for builderz-labs/mission-control, covering mission control, prerequisites, setup, run and docker.
helmo AGENTS.md
AGENTS.md instructions for arthurcoulston/helmo, covering agents — session routing for helmo, what kind of session is this? and universal expectations.
helmo CLAUDE.md
Claude Code instructions for arthurcoulston/helmo: The canonical apparatus for this repository is AGENTS.md — read it first; it routes every session (summoned roles load only their role file; dev sessions load DEV.md). This file exists only because Claude Code looks for CLAUDE.md. Substance never lives here.
Paddock AGENTS.md
AGENTS.md instructions for racecraft-lab/Paddock, covering paddock, openclaw node deployment notes, prerequisites, setup and run.
Paddock CLAUDE.md
Claude Code instructions for racecraft-lab/Paddock, covering paddock, openclaw node deployment notes, prerequisites, setup and run.
kungfu copilot-instructions.md
Copilot instructions for kungfu-systems/kungfu: Read and follow AGENTS.md before proposing or running changes. It is the shared agent router for this repository, including the canonical Shifu development and build entrypoint.