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/josephbartlett/codex-relay/agents-mdgit clone --depth 1 https://github.com/josephbartlett/codex-relayWrote 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/josephbartlett/codex-relay/agents-md)<a href="https://agentmods.dev/instructions/josephbartlett/codex-relay/agents-md"><img src="https://agentmods.dev/badge/instructions/josephbartlett/codex-relay/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 | $0.01348 | $0.01348 |
| Opus 5 | $0.00674 | $0.00674 |
| Sonnet 5 | $0.00270 | $0.00270 |
| Haiku 4.5 | $0.00135 | $0.00135 |
Grade A, and why
codex-relay 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 — 109 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Operating Guide
This repository is intended to be maintained by humans and coding agents. Agents must preserve repo hygiene, secrets hygiene, and release discipline from the beginning.
Prime Directive
Build Codex Relay as a production-shaped Slack control plane for local and self-hosted Codex execution. Favor durable, testable, documented changes over demo-only shortcuts.
Safety Rules
- Never print, commit, move, overwrite, or normalize
.envsecrets. - Do not edit
.envunless the user explicitly asks for a specific change. - Keep
.codex-slack/,dist/,node_modules/, logs, worktrees, tokens, and generated local runtime state out of git. - Do not use
danger-full-access,--yolo, or equivalent unsafe Codex execution defaults in product code. - Preserve workspace isolation: approved write tasks must run in session worktrees, not a user's active branch.
- Do not revert user changes unless explicitly requested.
Engineering Standards
- Use TypeScript with strict typechecking.
- Keep Slack gateway, orchestrator, local runner, shared types, infra, and docs boundaries clear.
- Prefer existing local patterns over new abstractions.
- Add abstractions only when they remove real duplication or define a durable boundary.
- Treat Slack messages, Codex prompts, command output, and diffs as sensitive data.
- Make state transitions idempotent where Slack retries or duplicate button clicks are plausible.
Custody-First Workflow
- Use Custody-First Orchestration for all non-trivial repository work.
- Create or activate a machine-readable packet in
docs/work-packets/before starting a meaningful implementation, release, security, governance, or visual-asset chunk. - Keep packet ownership narrow: one owner, exact paths or named concerns, explicit non-goals, acceptance criteria, verification, risks, and handoff expectations.
- Keep active/review/verify packet scopes disjoint. If two packets may touch the same files or concern, serialize the work.
- Run
npm run check:work-packetswhenever packet state changes. - Do not mark a packet
donewithout completed verification, evidence, handoff notes, reviewer/verifier, date, checks, and commit or artifact reference. - Record maintenance history in
docs/MAINTENANCE_AUDIT.mdand active status indocs/TASKS.md. - External Custody-First scaffold work is currently tabled. Do not edit any external scaffold repository unless the user explicitly reopens that project.
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 · 109 lines · 1,348 tokens per session scan A 318acf76a8c2
codex-relay AGENTS.md is an instructions file published in the GitHub repository josephbartlett/codex-relay (1 stars, last pushed 4mo ago), licensed MIT. It adds 1,348 tokens to every session, about $0.0067 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
martin-loop AGENTS.md
Instructions for Keesan12/martin-loop, covering martinloop agent operating rules, repository authority, mandatory internal-first sequence, prohibited actions and required internal health proof.
legalise AGENTS.md
Instructions for b1rdmania/legalise, covering legalise — agent context (agents.md), 1. the one-paragraph version, 2. what it's built to answer (the four questions), 3. stack (boring by design) and 4. the core concepts.
capsule AGENTS.md
Instructions for quantumpipes/capsule, covering agents.md: capsule, stack, commands, structure and style.
blueclaw AGENTS.md
Instructions for yeomyeonggeori/blueclaw, covering agents.md, core principles, single source of truth, working on this repository and branch names.
awesome-matematic-skills-pl AGENTS.md
AGENTS.md instructions for matematicsolutions/awesome-matematic-skills-pl, covering agents.md - awesome-matematic-skills-pl, project goal, matematic context (hard constraints), repo structure and build and test.
leashd CLAUDE.md
Instructions for vmehera123/leashd, covering claude.md, commands, install dependencies, run tests (single file / specific test / all) and run tests with coverage.