codex-skills-kit AGENTS.md

Repository instructions for a package of reusable coding-agent skills and the tools that validate and package them.

In plain words
What is it for?
Finding skill files, plugin metadata, examples, fixtures, and validation steps before editing or committing changes.
Why use it?
They give agents the project layout and working rules needed to change the repository without breaking its checks or conventions.

Instructions file for CodexOpenCode

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 instructions/takeoff69/codex-skills-kit/agents-md
Clone the repo
git clone --depth 1 https://github.com/TAKEOFF69/codex-skills-kit

Made for: Codex, OpenCode.

Per session 714 This file is loaded in full into every session.
When invoked 714 The same file — it is already loaded in full.
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 $0.00714 $0.00714
Opus 5 $0.00357 $0.00357
Sonnet 5 $0.00143 $0.00143
Haiku 4.5 $0.00071 $0.00071

Measured 2d ago against content hash 5acc3dc929bf, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

codex-skills-kit AGENTS.md 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.

AGENTS.md · 53 lines

How it starts

The opening of the file, as written. The whole thing — 53 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md – codex-skills-kit

Conventions for agents working in this repository. This is a content and tooling repo: reusable Codex workflow skills, plugin packaging, fixtures, and deterministic validation scripts. There is no application runtime.

What this repo is

A package of Codex skills covering prompt delegation, Git contention, gate engineering, honest verification, session closeout, and self-distillation. It also ships as a Codex plugin. Methodology is product; keep its depth while stripping project-specific details.

Layout

  • .codex-plugin/plugin.json – plugin manifest bundling all folders under skills/.
  • .agents/plugins/marketplace.json – repository marketplace catalog pointing at root plugin.
  • skills/<name>/SKILL.md – skill instructions and triggering frontmatter.
  • skills/<name>/agents/openai.yaml – optional Codex app metadata validated when present.
  • skills/<name>/scripts/ – optional deterministic helper with hermetic selftest when practical.
  • examples/<task-type>-example.md – worked prompts on fictional Acme Notes project.
  • fixtures/golden-prompts/<task-type>.md – compact canonical prompts validated against task registry.
  • task-types.json – source-of-truth mapping task types to skills, templates, examples, and fixtures.
  • scripts/validate_skills.py – skill, plugin, registry, fixture, release, link, and secret validator.
  • .github/workflows/validate-skills.yml – CI validation on pushes and pull requests.

Validate before committing

python scripts/validate_skills.py --selftest
python scripts/validate_skills.py
python skills/git-under-contention/scripts/contention_preflight.py --selftest

Green validation covers skill frontmatter, present Codex metadata, plugin and marketplace coherence, task registry, golden fixtures, local links and anchors, release metadata, secret patterns, and seeded validator failures.

Conventions

  • Style: en dashes (–) only in prose – never em dashes (—) or -- as punctuation. Syntactic -- in commands, SQL, and Markdown is fine.
  • Project-neutral: examples and skills reference only Acme Notes or generic placeholders – no private project names, infrastructure, paths, or incident identifiers.
  • Progressive disclosure: keep SKILL.md focused; move deep optional detail into directly linked references/.
  • Skill metadata: frontmatter must match validator; agents/openai.yaml default prompt must name $skill-name.
  • Persistent gates: prove new or changed gates with known-bad and known-good cases plus deterministic selftest.
  • Task-type changes move together: update codex-prompt, prompt templates, worked example, golden fixture, task-types.json, and README table in one change.
  • Plugin inventory changes move together: update plugin manifest description/default prompts, changelog version, README, and validation.
  • Diff discipline: every changed line traces to task; avoid drive-by formatting.

Read the full file on GitHub · 53 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.

  1. 2d ago First seen · 53 lines · 714 tokens per session scan A 5acc3dc929bf

Subscribe to this mod's changes

codex-skills-kit AGENTS.md is an instructions file published in the GitHub repository TAKEOFF69/codex-skills-kit (6 stars, last pushed 1mo ago), licensed MIT. It adds 714 tokens to every session, about $0.0036 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 instructions, from other repositories