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.
npx agentmods add skills/alessiomarcone/agent-fieldbook/codexnpx skills add alessiomarcone/agent-fieldbook --skill codexgit clone --depth 1 https://github.com/alessiomarcone/agent-fieldbookWrote 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/skills/alessiomarcone/agent-fieldbook/codex)<a href="https://agentmods.dev/skills/alessiomarcone/agent-fieldbook/codex"><img src="https://agentmods.dev/badge/skills/alessiomarcone/agent-fieldbook/codex.svg" alt="Measured on agentmods" 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 | $0.00072 | $0.00482 |
| Opus 5 | $0.00036 | $0.00241 |
| Sonnet 5 | $0.00014 | $0.00096 |
| Haiku 4.5 | $0.00007 | $0.00048 |
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 6d 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.
What it actually says
Codex
Turn an objective into a reviewable Codex task with the right context, authority, and completion checks.
References
- Always read
references/knowledge-base.mdfor the operating model. - Read
references/sources.mdbefore making a time-sensitive product claim. - For specialist authoring, route to
prompt,subagent,skill, orplugin.
Procedure
- Resolve the outcome, repository or folder context, surface, risk, and definition of done.
- Choose the surface: app for parallel local work and artifact workflows, CLI/IDE for interactive engineering, cloud for configured remote work, scheduled tasks for clock-driven recurring execution, or SDK/non-interactive mode for programmatic execution.
- Put durable repository commands, conventions, boundaries, and verification in the nearest applicable
AGENTS.md. - Add a focused skill for a repeated workflow and a custom agent only when isolation or a distinct configuration is valuable.
- Use the least authority that can complete the task; make external and irreversible actions explicit.
- Require relevant tests, review, and a concise final report of changes and remaining uncertainty.
- Check current official documentation for commands, fields, models, plans, and feature availability.
Output contract
Return:
- Surface and mechanism — one primary choice and why.
- Context setup — files, instructions, tools, and permissions.
- Ready task — a complete Codex prompt or configuration.
- Done when — observable checks.
- Reuse — what belongs in
AGENTS.md, a skill, custom agent, or plugin.
Boundaries
- Do not invent current commands, configuration, or availability.
- Do not absorb prompt, subagent, skill, or plugin authoring when a narrower skill matches.
- Do not request broader permissions than the task needs.
- Do not transfer Claude-only configuration into Codex.
- Answer in the user's language.
What ships with it
3 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.
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.
- 6d ago First seen · 43 lines · 72 tokens per session scan A b97b72c0f7da
codex is a skill published in the GitHub repository alessiomarcone/agent-fieldbook (5 stars, last pushed 5d ago), licensed MIT. It adds 72 tokens to every session and 482 once invoked, about $0.0004 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.
Other skills, from other repositories
acontext-installer
Install Acontext, Login & Init Acontext Project, Add Skill Memory to Agent.
lore-lint
Use when the user runs /lore:lore-lint — health-check the lore for index drift, ungrouped pages, group hygiene, dead wikilinks, duplicate titles, unresolved contradictions, schema violations, missing footnote attribution, missing concept pages and cross-references, knowledge gaps, and discard candidates. Fix…
lore-ingest
Use when the user runs /lore:lore-ingest — find files in the lore's raw/ inbox that are new or changed since their last ledger entry, distill or surgically update wiki pages by content type, update log.md, regenerate the index, and commit.
forge-writing
文献锚定型学术写作（IMRD 完整四段），三阶深挖 → 创新锚定 → 叙事线走通 → 逐段交互起草 → 综合审校。支持单章写作、全文编排、润色模式。当用户需要写论文段落、起草 Introduction/Discussion/Methods/Results/Abstract/Conclusion、润色已有草稿、讨论论文大纲或写作思路时触发——即使用户没有明确说"写作"，只要涉及论文内容产出（"帮我写一段引言""这个讨论怎么写""润色一下这段""帮我看看这个大纲"）都应使用此 skill。材料科学领域优化，深度整合 SciCrucible 知识库文献。.
forge-extract
通用实验数据提取框架 — 首次 Q&A 配置向导 → 小规模试点验证 → 全量检索提取入库。跨领域通用：材料科学、化学、物理等任意实验数据。当用户想从文献中构建实验数据集、提取结构化数据、建 ML 训练数据库时触发，即使用户没有说 /extract。.
forge-ingest
文献摄入 — 读取 PDF/文本，编译为结构化 wiki 页面（文献笔记 + 概念页 + 方法页），增量更新索引。当用户放入新PDF、说"摄入这篇论文"、"处理文献"、需要将PDF转为结构化知识时触发.