What is codeflow AGENTS.md?

A development guide for Codeflow, a project that uses a contract-first approach to keep code, documentation, tests, and repository checks aligned. It lists the required reading order and verification commands.

Bootstrapping the project, running CI and tests, checking workflows and Docker usage, scanning for secrets, auditing files, and maintaining documentation.

It helps an AI assistant make small, reviewable changes and catch security, workflow, documentation, and repository-hygiene problems before claiming success.

How xiaojiou176-open/codeflow configures Codex and OpenCode

Reuse

Borrowing it

Nothing to install: this file belongs to xiaojiou176-open/codeflow. 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/xiaojiou176-open/codeflow/main/AGENTS.md

Clone the repo

git clone --depth 1 https://github.com/xiaojiou176-open/codeflow

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.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/xiaojiou176-open/codeflow/agents-md/github.svg)](https://agentmods.dev/instructions/xiaojiou176-open/codeflow/agents-md)

Your own site

<a href="https://agentmods.dev/instructions/xiaojiou176-open/codeflow/agents-md"><img src="https://agentmods.dev/badge/instructions/xiaojiou176-open/codeflow/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.

Your own site · 80×15

<a href="https://agentmods.dev/instructions/xiaojiou176-open/codeflow/agents-md"><img src="https://agentmods.dev/badge/instructions/xiaojiou176-open/codeflow/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 6,303 This file is loaded in full into every session.

When invoked 6,303 The same file — it is already loaded in full.

Security scan D 2 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.

Model	Per session	Once invoked
Fable 5.1	$0.06303	$0.06303
Opus 5.5	$0.02521	$0.02521
Sonnet 5.5	$0.01261	$0.01261
Haiku 4.5	$0.00630	$0.00630

Measured 2mo ago against content hash 9423310cd698, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade D, and why

codeflow AGENTS.md scanned grade D with 2 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 2mo 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.

Reads agent configuration directoriesmediumAgent snooping

.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.

live-provider rule that resolves process env first and `~/.codex/config.toml`

Recursive force deletehighDestructive command

rm -rf with a variable or a broad path is one typo away from removing the wrong tree.

before the clean-room `rm -rf` sweep, plus quarantining stubborn dashboard

AGENTS.md · 442 lines

How it starts

The opening of the file, as written. The whole thing — 442 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS Guide

Mission

Work in Codeflow as a contract-first engineering agent:

keep diffs small and auditable
keep code and docs in sync
verify behavior with real commands before claiming success
never commit runtime caches, local secrets, logs, or generated noise

Canonical Read Order

README.md
docs/index.html
configs/repo_positioning.json
configs/docs_nav_registry.json
nearest local wrapper under apps/*/AGENTS.md or apps/*/CLAUDE.md

Key Commands

bootstrap: npm run bootstrap
local fast CI: npm run ci
local strict CI: npm run ci:strict
fast verification: npm run test:quick
main local gate: npm run test
host safety scan: npm run scan:host-process-risks
space audit: npm run space:audit
docker runtime audit: npm run docker:runtime:audit
hygiene: bash scripts/check_repo_hygiene.sh
workflow security: npm run scan:workflow-security
Trivy repo scan: npm run scan:trivy
closeout secret scan: npm run security:scan:closeout
truth split: npm run truth:triage
full pre-commit: pre-commit run --all-files
repo-owned scripts/*.py entrypoints must remain runnable through direct python3 scripts/<name>.py execution or bash scripts/run_governance_py.sh without relying on a pre-seeded repo-root PYTHONPATH
host-compatible pre-commit hooks that execute repo-owned scripts/*.py entrypoints must use the same wrapper path (or an equivalent python3 -B contract) so clean hook runs do not leave repo-local __pycache__ residue

Generated Governance Context

active CI layers: pre-commit, pre-push, hosted, nightly, manual
do not reintroduce a sixth CI/profile/workflow layer; move scheduled heavy checks to nightly and explicit high-cost verification to manual

trust flow: ci-trust-boundary -> quick-feedback -> hosted policy/core slices -> pr-release-critical-gates -> pr-ci-gate
hosted policy/core slices: policy-and-security, core-tests
untrusted PR path: quick-feedback -> untrusted-pr-basic-gates -> pr-ci-gate
protected sensitive lanes: workflow_dispatch -> owner-approved-sensitive -> ui-truth / resilience-and-e2e / release-evidence
canonical machine SSOT: configs/ci_governance_policy.json

Read the full file on GitHub · 442 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.

2mo ago First seen · 442 lines · 6,303 tokens per session scan D 9423310cd698

Subscribe to this mod's changes

codeflow AGENTS.md is an instructions file published in the GitHub repository xiaojiou176-open/codeflow (207 stars, last pushed 3mo ago), licensed MIT. It adds 6,303 tokens to every session, about $0.0252 per session on Opus 5.5. A static security scan graded it D with 2 findings (reads agent configuration directories, recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

ai-agents auditability control-plane developer-tools evidence fastapi

codeflow AGENTS.md next to 6 instructions from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
codeflow AGENTS.mdxiaojiou176-open/codeflow	Instructions file	207	3mo ago	MIT	D	6,303	Codex, OpenCode
runtm copilot-instructions.mdruntm-ai/runtm	Instructions file	303	yesterday	none	A	297	GitHub Copilot
vscode buildNext.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	9,719	GitHub Copilot
next.js AGENTS.mdvercel/next.js	Instructions file	143k	today	MIT	A	7,461	Codex, OpenCode
langchain AGENTS.mdlangchain-ai/langchain	Instructions file	149k	yesterday	MIT	A	4,680	Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	5,516	GitHub Copilot
vscode copilot-instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	3,093	GitHub Copilot

codeflow: Instructions file for Codex