codex-pipeline

codex-pipeline is a skill for Claude Code from V-Songbird/foundry. It costs 152 tokens per session (585 once invoked), scanned A, original, MIT.

A cross-model development workflow in which one session writes the specification and private verification tests, Codex implements the change, and a fresh review checks the result. Held-out tests are tests the implementer is not allowed to see beforehand.

In plain words
What is it for?
Use it when asking Codex or GPT to implement a coding task and you want a separate model review plus private verification before accepting the changes.
Why use it?
It adds independent review and adversarial checks instead of trusting the implementer’s summary or self-tests alone. It also verifies the change against the project’s normal tests, type checks, and lint rules.

Skill for Claude Code

Written for Claude Code: argument-hint in frontmatter. Also seen: mentions Codex.

Good fit Use it when asking Codex or GPT to implement a coding task and you want a separate model review plus private verification before accepting the changes.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/v-songbird/foundry/codex-pipeline
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.

Any agent
npx skills add V-Songbird/foundry --skill codex-pipeline
Clone the repo
git clone --depth 1 https://github.com/V-Songbird/foundry

Made for: Claude Code.

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 codex-pipeline

README.md
[![agentmods](https://agentmods.dev/badge/skills/v-songbird/foundry/codex-pipeline/github.svg)](https://agentmods.dev/skills/v-songbird/foundry/codex-pipeline)
Your own site
<a href="https://agentmods.dev/skills/v-songbird/foundry/codex-pipeline"><img src="https://agentmods.dev/badge/skills/v-songbird/foundry/codex-pipeline/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 codex-pipeline

Your own site · 80×15
<a href="https://agentmods.dev/skills/v-songbird/foundry/codex-pipeline"><img src="https://agentmods.dev/badge/skills/v-songbird/foundry/codex-pipeline.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 152 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 585 The whole file, excluding the scripts and references it only reads on demand.
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 $0.00152 $0.00585
Opus 5 $0.00076 $0.00293
Sonnet 5 $0.00030 $0.00117
Haiku 4.5 $0.00015 $0.00059

Measured today against content hash 4d081fd6000b, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

codex-pipeline 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 today.

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.

.claude/skills/codex-pipeline/SKILL.md · 54 lines

What it actually says

Codex pipeline

Cross-model build-and-review. This session architects, Codex implements, a fresh context reviews. The fresh review is the point: a second model with different blind spots catches what the implementer and the tests both miss.

1. Architect

  • Read the code the task touches.
  • Write a self-contained spec file: the contract, required error types, and an "Adversarial notes" section listing the edge cases you'd attack.
  • Write a held-out verification suite yourself, named distinctly (e.g. verify.test.js). Codex must never see it.

2. Implement

Call mcp__codex__codex with:

  • prompt: implement from the spec file, write and run its own selftest, and do NOT open the held-out suite (name it explicitly).
  • cwd set to the working directory, sandbox: workspace-write, approval-policy: never.

Save the returned threadId. Codex returns only a one-paragraph summary — trust the files on disk, never the summary.

3. Verify

Run the held-out suite plus the project's real gates (tests, types, lint) yourself, in this session.

4. Fresh review

Spawn a general-purpose Agent with no pipeline context: give it only the spec path and the implementation path, read-only intent. Require it to reproduce every suspected defect before reporting, and to return APPROVE or REQUEST-CHANGES with repro input and actual-vs-expected output per defect.

5. Fix

Send confirmed defects — with their repros — to mcp__codex__codex-reply on the saved threadId. Then re-run step 3.

6. Close

  • At most two review→fix rounds; hand anything still open to the user.
  • Report: what was built, verification results, what the fresh review caught.
  • Never commit without the user's explicit go.
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. today First seen · 54 lines · 152 tokens per session scan A 4d081fd6000b

Subscribe to this mod's changes

codex-pipeline is a skill published in the GitHub repository V-Songbird/foundry (4 stars, last pushed today), licensed MIT. It adds 152 tokens to every session and 585 once invoked, about $0.0008 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 2026-09-10.

Related

Other skills, from other repositories