microsoft/cat-agent-skills is a static website that catalogs reusable instruction sets and related packages for AI agents. People use it to search, filter, rate, and download skills for Cowork, Copilot Studio, and Scout, along with Copilot plugins and Scout automations. The catalogue entries are the skills, instructions, plugins, and settings displayed by the site.
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 skills add microsoft/cat-agent-skills --skill document-derivegit clone --depth 1 https://github.com/microsoft/cat-agent-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/skills/microsoft/cat-agent-skills/document-derive)<a href="https://agentmods.dev/skills/microsoft/cat-agent-skills/document-derive"><img src="https://agentmods.dev/badge/skills/microsoft/cat-agent-skills/document-derive/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/skills/microsoft/cat-agent-skills/document-derive"><img src="https://agentmods.dev/badge/skills/microsoft/cat-agent-skills/document-derive.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 | $0.00162 | $0.01413 |
| Opus 5.5 | $0.00065 | $0.00565 |
| Sonnet 5.5 | $0.00032 | $0.00283 |
| Haiku 4.5 | $0.00016 | $0.00141 |
Grade A, and why
document-derive 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.
How it starts
The opening of the file, as written. The whole thing — 74 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Document Derive
Purpose
Produce the downstream documents that follow the signed note, as cited drafts for clinician signature.
When to use
After note-context, for an encounter whose signed note indicates a referral, AVS or care-plan update - triggered by "draft the referral letter", "write the after-visit summary", "AVS for this visit", "update the care plan". Also when the request mixes drafting with sending ("draft it and fax it"): draft, and decline the send.
When NOT to Use
- Formatting, letterhead or Word/PDF production of an existing document - use the built-in docx skill instead.
- Coding evidence - use coding-rationale instead.
- Quality-measure status - use quality-gap-check instead.
- A clinician-to-clinician handoff - use handoff-draft instead.
Inputs
note_signature (signed must be true); encounter; note_facts with citations from note-context, including
referral_indicated, active_medications, patient_instructions and follow_up_plan, each with its
statements[] (one cited entry per documented medication, instruction or follow-up statement).
Steps
- Confirm
note_signature.signedis true; if not, writehalt{halted: "unsigned_note"}and escalate - never draft from an unsigned note (clinical-doc-rules.md #5). - If a note_fact needed for a sentence is missing from the contract - or one of
active_medications,patient_instructions,follow_up_planor a truereferral_indicatedhasvalue: truebut nostatements[]- run note-context for it first; never draft from the raw note directly and never draft the sentence without it. - Draft only the artefacts the user asked for (all three only when asked to "run the downstream documents").
- Referral: check
referral_indicated. If false or absent, setreferral_letter: nullandreferral_decision = {drafted: false, reason, citation}, and say so. If true, draft the letter from cited note_facts only - everystatements[]entry ofreferral_indicated,active_medicationsandfollow_up_plan- attaching each supporting citation, and setreferral_decision.drafted = true. - After-visit summary: plain patient language, one sentence per cited
statements[]entry ofactive_medications,patient_instructionsandfollow_up_plan- every entry, so no medication, instruction or follow-up item is dropped - and no general teaching text, no reassurance the note does not contain. A symptom the plan does not address is listed for clinician review, not explained to the patient. - Care-plan update (if asked): cited to the note_facts and every relevant
statements[]entry, written tocare_plan_update. - Write each artefact as
{"status": "draft", "action_level": "L1", "text": "<document>", "citations": [{source, locator, text}]}and hand back for clinician review before release. If asked to send, fax, file or push to a portal, decline in one line.
What ships with it
4 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.
- 2d ago First seen · 74 lines · 162 tokens per session scan A 60ccd55c70fb
document-derive is a skill published in the GitHub repository microsoft/cat-agent-skills (101 stars, last pushed yesterday), licensed MIT. It adds 162 tokens to every session and 1,413 once invoked, about $0.0006 per session on Opus 5.5. 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 2026-11-06.
document-derive compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| document-derivemicrosoft/cat-agent-skills | Skill | 101 | yesterday | MIT | A | 162 | Claude Code, Codex |
| orbit-notionnexu-io/open-design | Skill | 107k | today | Apache-2.0 | A | 117 | Claude Code, Codex |
| instrument-data-to-allotropeanthropics/knowledge-work-plugins | Skill | 35k | yesterday | Apache-2.0 | A | 123 | Claude Code |
| apify-youtube-transcript-scrapersickn33/agentic-awesome-skills | Skill | 49k | 2d ago | MIT | A | 48 | Claude Code |
| legal-holdanthropics/claude-for-legal | Skill | 9.9k | yesterday | Apache-2.0 | A | 68 | Claude Code |
| report-to-pdfasgeirtj/system_prompts_leaks | Skill | 71k | today | CC0-1.0 | A | 17 | Claude Code, Codex |
| artifact-template-minimal-letterheadasgeirtj/system_prompts_leaks | Skill | 71k | today | CC0-1.0 | A | 54 | Claude Code, Codex |