Codex: five config rules and a five-check script for .codex/config.toml (free sample)
by PaperChaseWebb
From the Codex bundle: five rules for .codex/config.toml (keep the sandbox on, approval on-request, network off by default, pin the model, pin servers and keep keys out), and a read-only script that checks one config file for them.
- Price
- Free
- Difficulty
- Beginner
- Setup
- 5 min
- Format
- .md file
codex-five-config-rules-and-a-five-check-script-free-sample.md · 7 KB · direct download, no signup
What you end up with
In a few minutes you can see whether your Codex config runs without a sandbox, never asks, has network on, or holds a token in plain text.
What you get
- Five config rules
- cx_mini.py: five read-only checks on one config.toml
- One prompt
- Worked example: 6 findings across the five checks, then 0
Who it is for
Anyone on Codex who wants a quick check of their config first.
Works with
Solid chips: the agent is named in this listing's own tools list. Dashed chip: Plain Markdown, so it also pastes into other agents such as Grok Bot, Hermes, OpenClaw or Muse. We list those as compatible by format only, not as tested.
Tools needed
Cost to run: Free.
Time to result: About five minutes to a first result
Sample prompts, free to use
Read .codex/config.toml and ~/.codex/config.toml. Tell me the sandbox mode, approval policy, network setting and model you are actually running with, and which file each one came from. Change nothing.
The rules and the script are in the file.
Free preview of the file
# Codex: five config rules and a five-check script for .codex/config.toml (free sample)
Version 1.0, 2026-10-29. By PaperChaseWebb. For developers who use Codex every day, alone or next to Claude Code, Hermes or Grok Bot. One Markdown file and one Python script, standard library only (Python 3.11 or newer, for tomllib). The script was tested on its worked example (6 findings across its five checks, then 0) and on a missing file (exit 2). It reads one config file; it was not run inside Codex or any other harness.
Most Codex setups that go wrong go wrong in the same file. The sandbox was set to danger-full-access to get past one prompt and never set back. approval_policy is never, so nothing stops. Network is on for every task. No model is pinned, so a default change switches it under you. An MCP server is installed at @latest with its token written into the config. This sample gives you the five rules that cover those, and a script that checks them in one config.toml.
## Install in one line
Save this page as FREE-SAMPLE.md, then:
```
python3 -c "import re;t=open('FREE-SAMPLE.md').read();open('cx_mini.py','w').write(re.search(r'~~~python\n(.*?)~~~',t,re.S).group(1))"
python3 cx_mini.py .codex/config.toml
```
## Five config rules
1. **Keep the sandbox on.** sandbox_mode is read-only or workspace-write. danger-full-access is for a throwaway machine, not your project.
2. **Let Codex ask.** approval_policy "on-request". never means nothing stops it, and "untrusted" is no lonTest record
- Date
- Oct 29, 2026
- What was run
- cx_mini.py on the worked example, a fixed file, a missing file, and the copy taken from the sample file.
- Result
- 6 findings (one check found two), then 0; exit 2 on a missing file; the copy behaved the same.
- Not run
- Codex or any other harness.
Copied from this item's dated test notes. What was not run stays in.
Worth knowing
- The sample checks one config file against five rules. The full Codex Setup Runbook, gate, request approvals by hash, source and vetting records and the fifteen checks are in the full bundle.
- On 2026-10-29 cx_mini.py gave 6 findings on the worked example (one check found two), then 0 on the fixed file, and exited 2 on a missing file; the copy extracted from the sample file behaved the same. It was not run inside Codex or any other harness.
Questions
Does it send anything anywhere?
No. It reads one file and opens no network connection.
What is in the full bundle?
The Codex Setup Runbook with fifteen checks and owner approval tied to each request's hash, plus Spec-First Build Docs, Two-Agent Build and Test Loop, Planner Over Cheap Models and Token Saver in full. $69, against $101 for the four bought separately.
Details
Version 1.0 · Updated Oct 29, 2026 · Markdown version of this page
License: Personal and commercial use. Download terms
Related
Codex bundle: spec, plan, build, test and keep the context small, with a setup runbook a script checks
Spec-First Build Docs, Two-Agent Build and Test Loop, Planner Over Cheap Models and Token Saver in full, tied into one Codex workflow by a new Codex Setup Runbook. cx_check.py reads AGENTS.md, .codex/config.toml and .agents/skills and names fifteen kinds of problem: danger-full-access, approval set to never, network left on, keys Codex ignores in a project config, an AGENTS.md past the 32 KiB limit, an unpinned model, servers at @latest, tokens in env, skills that cannot trigger, and third-party pieces nobody vetted. cx.py gates ten risky setup changes, and each waits for the owner's yes typed at their terminal and tied to that request's sha256.
Claude Code: five settings rules and a five-check script for .claude/settings.json (free sample)
From the Claude Code bundle: five rules for .claude/settings.json (bypass stays off and switched off, deny reading .env, allow named commands not whole tools, git push asks, hooks run scripts kept in the project), and a read-only script that checks one settings file for them.
Spec-First Build Docs: PRD, design system, architecture note, agents guide and readiness check
A Markdown pack of the four documents to write before an agent builds anything, sized to the stakes: a PRD an agent can build from without inventing scope, the smallest design system that holds, a one-page architecture note with reversal costs, an agents guide with a project tree for Claude Code, Codex and Cursor, a weekly staleness pass and a nine-area readiness check.
Token saver: keep long command output out of your coding agent's context
Two shell scripts and one rule line that make Claude Code, Codex, Cursor and other agents read a one-page summary of builds, tests and searches instead of thousands of lines.