codex-autonomous-dev

A workflow layer for running long, mostly unattended Codex development sessions with saved state, task queues, isolated workspaces, retries, and verification records.

In plain words
What is it for?
It is for overnight or hands-off software development, including queued tasks, context handoffs, safe retries, resource limits, and evidence that work was checked.
Why use it?
It reduces interruptions during extended work while keeping tasks scoped to a selected project and goal.

Skill for Claude CodeCodex

Install

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.

agentmods
npx agentmods add skills/dear-ded/nightpilot-codex/codex-autonomous-dev
Any agent
npx skills add Dear-Ded/nightpilot-codex --skill codex-autonomous-dev
Clone the repo
git clone --depth 1 https://github.com/Dear-Ded/nightpilot-codex

Made for: Claude Code, Codex.

Per session 122 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,791 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00122 $0.01791
Opus 5 $0.00061 $0.00896
Sonnet 5 $0.00024 $0.00358
Haiku 4.5 $0.00012 $0.00179

Measured 2d ago against content hash 597ecd753d81, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

codex-autonomous-dev 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.

skills/codex-autonomous-dev/SKILL.md · 169 lines

How it starts

The opening of the file, as written. The whole thing — 169 lines — stays where its author put it; the contents beside it link to each section on GitHub.

NightPilot / 夜航员

Purpose

NightPilot is a lightweight operating layer around original Codex. It helps Codex run long development sessions with fewer interruptions by using durable local state, permission hardening where Codex supports it, task queues, branch/worktree isolation, retry classification, context handoffs, resource guards, and verification evidence.

NightPilot is Codex-coupled. It augments a Codex conversation, Codex CLI session, or explicit user-started Codex project context. It is not a standalone agent, daemon, scheduler, or replacement Codex runtime, and it must not initiate unrelated work outside the selected project and goal.

It does not bypass Codex internals, real account authorization, payment, CAPTCHA, missing secrets, or platform outages.

Critical hallucination guard: when the network fails, do not rewrite network configuration, edit DNS/proxy/firewall/router settings, restart or reopen the parent Codex app, reinstall Codex, create a replacement Codex, or attempt to rewrite Codex. Classify the failure, retry with bounded backoff when safe, park the affected task if needed, and continue independent local work.

Before changing this workflow, read NIGHTPILOT_REQUIREMENTS_LOCK.md. That file is the current source of truth for product direction and acceptance criteria.

Trigger Policy

Use this skill whenever the user asks for long-running, unattended, full-auto, autonomous, goal-mode, overnight, hands-off, or old-workflow takeover development.

Common trigger phrases include:

  • English: long-running, unattended, full-auto, autonomous, goal mode, overnight, hands-off, keep running, no approval prompts, take over this project, start now.
  • Brand/plugin names: NightPilot, 夜航员, $codex-autonomous-dev, codex-autonomous-dev.
  • Chinese: 长时间开发, 无人开发, 目标模式, 全自动, 自动跑, 自动开发, 自动创建分支, 撒手不管, 睡觉也能跑, 人不在也继续, 不要一直申请权限, 不要中断, 网络波动不中断, 授权不中断, 接管当前项目, 老对话接管, 直接开始全自动开发.

When triggered, do not merely explain the workflow. Inspect the current project, initialize or reuse .codex-autonomous/, harden supported Codex permission paths, adopt the goal, audit readiness, and start the run loop when the user asked for execution.

Read the full file on GitHub · 169 lines

Files

What ships with it

4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 2d ago First seen · 169 lines · 0 tokens per session scan A 597ecd753d81

Subscribe to this mod's changes

codex-autonomous-dev is a skill published in the GitHub repository Dear-Ded/nightpilot-codex (1 stars, last pushed 1mo ago), licensed MIT. It adds 122 tokens to every session and 1,791 once invoked, about $0.0006 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.

Related

Other skills, from other repositories

autogpt-agents

Autonomous AI agent platform for building and deploying continuous agents. Use when creating visual workflow agents, deploying persistent autonomous agents, or building complex multi-step AI automation systems.

davila7/claude-code-templates · 39 tokens

yao-meta-skill

Create, improve, or evaluate an existing skill from workflows, prompts, SOPs, scripts. Use for migration/release/package, routing, evals/tests, install/trust checks, 优化已有 skill, 补 trigger 评测. Exclude summary/translation/docs, brainstorming, one-off prompts, copy-only edits, outline-only requests.

yaojingang/yao-meta-skill · 76 tokens

biomed-skill-creator

Create new biomedical skills or improve existing ones for the OpenBioMed toolkit. Use this skill when: (1) Creating a new skill from scratch, (2) Capturing a workflow as a reusable skill, (3) Automating a biomedical task, (4) Improving an existing skill. This skill guides through an interactive process: define intent…

PharMolix/OpenBioMed · 89 tokens

mcp-builder

Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).

activepieces/activepieces · 61 tokens

agent-workflow-playbook

AI Agent Workflow & Skill Architecture Guide — turn expert work into measurable, reusable agent systems. Covers workflow discovery, skill decomposition, harness design, evaluation, human escalation, observability, cost control, and multi-agent orchestration. Includes a measured marketing-delivery case: 15 people × 3–4…

Gingiris-1031/gingiris-skills · 83 tokens

next-task

Run one unattended IMPLEMENTATION iteration of the autonomous value-creation loop — steward any in-flight PR, fix interrupts (red CI / security / human bugs), or else build ONE queued idea issue on a branch off auto-dev and open a PR that squash-merges on green CI. Ideation lives in the next-idea skill; this skill…

breaking-brake/cc-wf-studio · 107 tokens