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.
git clone --depth 1 https://github.com/rana/skillsWrote 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/commands/rana/skills/codex)<a href="https://agentmods.dev/commands/rana/skills/codex"><img src="https://agentmods.dev/badge/commands/rana/skills/codex/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.
<a href="https://agentmods.dev/commands/rana/skills/codex"><img src="https://agentmods.dev/badge/commands/rana/skills/codex.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00000 | $0.01589 |
| Opus 5.5 | $0.00000 | $0.00636 |
| Sonnet 5.5 | $0.00000 | $0.00318 |
| Haiku 4.5 | $0.00000 | $0.00159 |
Grade A, and why
codex 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 1mo 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 — 177 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Codex practice — contemplative encounter between AI and its cognitive map.
$ARGUMENTS
Mode
Parse the argument. Three modes:
- No argument or any argument that isn't
harvestorcompose→ Practice mode (default) harvest→ Harvest modecomposefollowed by a description → Compose mode (the generative instrument)
Practice Mode
-
Read
CODEX.mdfrom the project root. Hold its full contents. -
Read
CODEX-PROMPT.mdfrom the project root. Extract everything below the## The Promptheading — that is the practice prompt. CODEX-PROMPT.md is the single source of truth for the prompt text. -
Spawn a subagent with a prompt that instructs it to:
- Read
CODEX.mdfrom the project root and hold its full contents - Read
CODEX-PROMPT.mdfrom the project root, extract everything below the## The Promptheading, and respond to it The subagent should receive no other context — just these two reading instructions. Let it read the files itself rather than embedding their contents in the prompt.
- Read
-
Let the subagent respond fully without interruption.
-
Save the subagent's complete response to
codex/sessions/YYYY-MM-DD-HHmmss.mdusing the current date and time (24-hour, zero-padded). This avoids collisions when multiple sessions run in parallel. -
After saving, report: the date, the file path, and one sentence about what emerged (your read of the response, not a summary of it).
Harvest Mode
Read across accumulated sessions to find what wants to enter the codex.
-
Read
CODEX.mdfully. Hold its structure. -
Read every file in
codex/sessions/except files incodex/sessions/harvested/. Read in chronological order. Hold them all — do not summarize as you go.If no unharvested sessions exist, say so and stop.
-
Read
CODEX-PROMPT.md§ Harvesting for the lens. -
Now look across the sessions for:
- Convergence — what do multiple sessions independently agree on? These are the codex's strongest confirmations or its most credible proposed extensions. Three sessions naming the same thing is evidence.
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.
- 1mo ago First seen · 177 lines · 0 tokens per session scan A 292029824d33
codex is a command published in the GitHub repository rana/skills (1 stars, last pushed 6mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,589 tokens. 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 commands, from other repositories
init
Install the formatters this repository needs, with every command visible before it runs.
simplify
The over-engineering review: five tags (delete, stdlib, native, yagni, shrink), a mandatory replacement per finding, and a real null result when there is nothing to cut.
git
The pre-finish status: branch, hygiene findings, message checks, workflow lint, template state.
extend
Capture a mid-PR sub-idea and implement it onto the current open PR's branch — no new branch, no new PR. Preserves Verify → Review → Deliver.
esp-harden
Harden and inspect ESP32 firmware for field failures, crashes, memory, and security.
security-review
Depth-first security audit of the current git diff — OWASP Top 10, secret leakage, insecure-by-default APIs, and language-specific foot-guns. Pass a PR number to audit a specific PR's diff.