Wakeflow: Instructions file for Codex

AGENTS.md

Wakeflow AGENTS.md is an instructions file for Codex, OpenCode from GxFn/Wakeflow. It costs 3,146 tokens per session, scanned A, original, MIT.

Repository instructions for maintaining Wakeflow, a plugin that provides coding-agent guidance and tools. They define the project scope, safety rules, ownership boundaries, and Git practices.

In plain words
What is it for?
Use them when modifying Wakeflow to inspect existing changes, stay within scope, avoid destructive Git operations, and follow its authorization rules.
Why use it?
They help prevent changes from spilling into product repositories or accidentally overwriting other contributors’ work.

Instructions file for CodexOpenCode

Written for Codex and OpenCode: the file is AGENTS.md. Also seen: mentions CLAUDE.md; mentions Claude Code; mentions AGENTS.md.

This is GxFn/Wakeflow's own configuration. It tells Codex and OpenCode how to work on Wakeflow itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything Wakeflow configures →

Reuse

Borrowing it

Nothing to install: this file belongs to GxFn/Wakeflow. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/GxFn/Wakeflow/main/AGENTS.md
Clone the repo
git clone --depth 1 https://github.com/GxFn/Wakeflow

Made for: Codex, OpenCode.

Wrote 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.

agentmods badge for Wakeflow AGENTS.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/gxfn/wakeflow/agents-md/github.svg)](https://agentmods.dev/instructions/gxfn/wakeflow/agents-md)
Your own site
<a href="https://agentmods.dev/instructions/gxfn/wakeflow/agents-md"><img src="https://agentmods.dev/badge/instructions/gxfn/wakeflow/agents-md/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for Wakeflow AGENTS.md

Your own site · 80×15
<a href="https://agentmods.dev/instructions/gxfn/wakeflow/agents-md"><img src="https://agentmods.dev/badge/instructions/gxfn/wakeflow/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 3,146 This file is loaded in full into every session.
When invoked 3,146 The same file — it is already loaded in full.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.03146 $0.03146
Opus 5.5 $0.01258 $0.01258
Sonnet 5.5 $0.00629 $0.00629
Haiku 4.5 $0.00315 $0.00315

Measured 5d ago against content hash a336959b8467, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

Wakeflow 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.

Origin

Copies of this mod

1 near-identical copy found in the catalogue:

AGENTS.md · 218 lines

How it starts

The opening of the file, as written. The whole thing — 218 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Wakeflow Source Repository Instructions

This checkout is the development repository for the Wakeflow Codex and Claude Code plugin artifacts. Work here maintains Wakeflow itself; it is not an installed controller workspace and it does not own the product repositories used to test Wakeflow.

The generated plugin artifacts under plugins/ carry the skills, commands and READMEs shipped to installed controller workspaces. They describe an installed workspace's rules, not this repository's. Do not mistake their controller-role restrictions for a prohibition on authorized Wakeflow source maintenance here.

Scope And Safety

  • Follow the user's requested scope. Do not expand a Wakeflow change into an Alembic product change or another external repository.
  • Preserve pre-existing and unrelated working-tree changes. Inspect git status --short --branch and the relevant diff before editing.
  • Do not use destructive Git operations such as git reset --hard, discard another contributor's changes, or rewrite history unless the user explicitly authorizes that exact operation.
  • Do not commit, push, tag, publish, or refresh an installed plugin cache unless the user explicitly requests that action. Authorization for one of these actions does not imply authorization for the others.
  • Use only explicitly designated disposable workspaces for destructive runtime tests. Never treat a real product repository as a Wakeflow fixture.
  • Keep secrets, tokens, private session or thread identifiers, local absolute paths, and machine-specific cache paths out of committed source, fixtures, and docs.

Source Ownership

  • src/, tooling/ and tests/ are the only hand-written code. src/ is the runtime in six layers — foundation → contracts → kernel → capabilities / governance / configuration / workspace → hosts → entrypoints — and npm run check:architecture enforces the direction.
  • plugins/codex-wakeflow/ and plugins/claude-code-wakeflow/ are generated by tooling/artifacts/build-plugin-artifacts.ts. Never edit them by hand: change the source, run npm run build:artifacts:committed, and let npm run build:check prove the committed bytes match a fresh build.
  • assets/agent-text/ is the single source of the skills, commands and READMEs in both artifacts. Host differences are the twelve placeholders filled by src/hosts/<host>/<host>-agent-text-profile.ts; the build fails on an unregistered placeholder or an unused value.
  • Host-specific behavior lives only in src/hosts/<host>/: profiles, the hook fragment, the agent-text value table, maintenance execution, Claude's settings and status-line assets. Shared code consumes values a host profile supplies; it must not infer Codex-versus-Claude behavior through ad hoc host checks.
  • The Claude Code host has two window transports, tmux and desktop sessions, and both are first-class and maintained together; neither replaces the other. A change to the window transport layer (launch instructions, the helper asset, binding observations, liveness observations, the maintenance asset set, handover text) lands on both transports in the same batch, or registers the gap in docs/references/open-items.md (section Y for tmux, section X for desktop) with the batch that restores parity. Shared code consumes transport values from the host profile and never checks the transport name ad hoc. The plan is docs/reviews/2026-10-30-claude-host-dual-transport-plan.md.
  • src/contracts/schemas/ holds the portable JSON Schemas; src/contracts/generated/ is derived by npm run schema:build and checked for drift by npm run schema:check. Do not hand-edit generated files.
  • assets/release/version.json is the only version input. The two plugin package.json files, the two plugin manifests and the Claude marketplace entry must agree with it.
  • Treat plugin cache directories as installed outputs, never as source. Modify this checkout first and refresh a cache only from a validated plugin artifact when the user asks for it.
  • tooling/cli.ts is the maintainer CLI (npm run wf -- verify|doctor|lab| fault|live|capture|ci|timings), documented in docs/references/maintainer-tools.md. It is development support only: it never enters the artifacts, owns no Demand state, and its lab, live and capture receipts under .build/ carry local paths and stay uncommitted.
  • .github/workflows/verify.yml runs the same wf verify gate|artifact profiles on Ubuntu and macOS and uploads only the bounded ci export report. A green local gate is not a remote run; say which one ran.
  • wakeflow-architecture-atlas/ is a separate read-only map subproject with its own npm run check in that directory; maps/ is its canon. It is not part of the root gate and must not be cited as runtime authority.

Read the full file on GitHub · 218 lines

Changes

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.

  1. 5d ago Changed · +10 lines · +161 tokens per session a336959b8467
  2. 10d ago Changed · +5 lines · +96 tokens per session 3aeec32609c6
  3. 14d ago Changed · +2 lines · +21 tokens per session 3e3d2be888cd
  4. 15d ago Changed · +14 lines · +260 tokens per session 91461d4c29f0
  5. 17d ago Changed 63c5e2b94e8a
  6. 18d ago Changed · +11 lines · +246 tokens per session 0d7294aa193f
  7. 19d ago Changed · +39 lines · +652 tokens per session f17cd0929ab5
  8. 1mo ago Changed · +20 lines · +341 tokens per session 42cc3ccfddb8
  9. 2mo ago Changed · +5 lines · +110 tokens per session 5930de694e3d
  10. 2mo ago First seen · 112 lines · 1,259 tokens per session scan A e97cc156bc14

Subscribe to this mod's changes

Wakeflow AGENTS.md is an instructions file published in the GitHub repository GxFn/Wakeflow (5 stars, last pushed yesterday), licensed MIT. It adds 3,146 tokens to every session, about $0.0126 per session on Opus 5.5. 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.

Related

Wakeflow AGENTS.md compared with instructions from other repositories

Wakeflow AGENTS.md next to 6 instructions from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
Wakeflow AGENTS.mdGxFn/WakeflowInstructions file5yesterdayMITA3,146Codex, OpenCode
runtm copilot-instructions.mdruntm-ai/runtmInstructions file303yesterdaynoneA297GitHub Copilot
vscode buildNext.instructions.mdmicrosoft/vscodeInstructions file195ktodayMITA9,719GitHub Copilot
next.js AGENTS.mdvercel/next.jsInstructions file143kyesterdayMITA7,461Codex, OpenCode
langchain AGENTS.mdlangchain-ai/langchainInstructions file149ktodayMITA4,680Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscodeInstructions file195ktodayMITA5,516GitHub Copilot
vscode copilot-instructions.mdmicrosoft/vscodeInstructions file195ktodayMITA3,093GitHub Copilot