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 anylegal-ai/anylegal-oss --skill draftgit clone --depth 1 https://github.com/anylegal-ai/anylegal-ossWrote 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/anylegal-ai/anylegal-oss/draft)<a href="https://agentmods.dev/skills/anylegal-ai/anylegal-oss/draft"><img src="https://agentmods.dev/badge/skills/anylegal-ai/anylegal-oss/draft/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/anylegal-ai/anylegal-oss/draft"><img src="https://agentmods.dev/badge/skills/anylegal-ai/anylegal-oss/draft.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 | $0.00066 | $0.05081 |
| Opus 5 | $0.00033 | $0.02540 |
| Sonnet 5 | $0.00013 | $0.01016 |
| Haiku 4.5 | $0.00007 | $0.00508 |
Grade A, and why
draft scanned grade A with 1 finding 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 12d 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.
Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
- `docx` (docx-js v9.5.1) is globally installed in the Node sandbox. Do not probe with `npm list docx`, `subprocess.run`, or similar — just `require('docx')`. How it starts
The opening of the file, as written. The whole thing — 480 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Document Drafting (docx-js, from-scratch only)
When to Use
Use this skill when:
- The user asks to draft a new document, clause, or section from scratch
- The user says /draft
- The user asks to "create", "write", "generate", or "prepare" legal content
Do NOT use this skill for:
- Filling placeholders in an existing template (e.g.
[Disclosing Party]→Acme Corp) — that is an edit task. InvokeSkill(skill="docx-editing")and useinstantiate_template(oredit_documentfor per-placeholder review). - Editing or restructuring an existing DOCX — also
docx-editing.
Critical rules (read first)
These rules constrain how you call the tool. Following them produces documents that open cleanly in Word, Google Docs, and LibreOffice; ignoring them produces invalid output.
| # | Rule | Why |
|---|---|---|
| 1 | DOCX creation goes through run_code(language="node") calling docx-js. Not python-docx. Not pandoc. Not lxml. |
docx-js is the only library with reliable APIs for footnotes, internal hyperlinks, TableOfContents, page numbers in headers/footers, and positional tabs. |
| 2 | One document = one run_code(language="node") call. Write the complete document in a single call — no skeleton-then-fill, no part-by-part assembly. |
docx-js has no Document.load() API; a second call to the same filename silently overwrites the first. |
| 3 | Bullets are produced via numbering config, not inline characters. Never write new TextRun("• Item") — use LevelFormat.BULLET with a numbering block on the Document. See "Lists" below. |
Inline bullet glyphs render inconsistently across viewers and break list semantics. |
| 4 | Filenames must be distinct logical names. Avoid _Part1 / _Final / _Draft / _v2 / _Copy suffixes. One logical document = one filename. |
Versioning suffixes signal split-document anti-patterns and confuse downstream document management. |
| 5 | Multi-document sets: todo_write first. If the user asks for a "set" / "series" / "package", your first tool call is todo_write with one item per document. After each document is saved, mark it completed and continue with the next item — in the same turn. |
Without an explicit todo list, multi-doc requests get partially completed and abandoned. |
| 6 | Content completeness. The saved document must be complete, not a skeleton: schedules referenced in the body must exist, defined terms must appear in the definitions clause, both parties' signature blocks for agreements, witness blocks for deeds. | A skeleton wastes the user's time — they have to ask again to fill the gaps. |
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.
- 12d ago First seen · 480 lines · 66 tokens per session scan A c11f655a8943
draft is a skill published in the GitHub repository anylegal-ai/anylegal-oss (11 stars, last pushed 4mo ago), licensed MIT. It adds 66 tokens to every session and 5,081 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
html-ppt-zhangzara-stencil-tablet
A workplace-safety compliance review for a manufacturing regulator — findings, the evidence chain, and the corrective mandate. Built as a decision-grade policy briefing deck for regulator, plant leadership.
treatment-plans
Format and structurally validate local treatment-plan documentation after clinical decisions have already been supplied and verified by authorized licensed professionals. Use for source traceability, clinician-authored intervention records, goals and checkpoints, shared-decision records, reconciliation handoffs, and…
handbook-updates
Diff a proposed handbook change against the current version, flag ripple effects and state supplement impacts. Use when user says "update the handbook", "add this to the handbook", "handbook change", or has a policy ready for insertion.
software-copyright-materials
A workflow for creating Chinese application materials for software copyright registration from a real software project.
iflytek-contract-intelligence-review
A contract-review workflow for scanned or digital agreements. It can recognize document text, examine clauses, detect risks, check compliance, and create translation summaries; its output is for review and is not formal legal advice.
legal-writer
Legal document drafting -- contracts, memos, briefs, complaints, demand letters, opinions, discovery, settlements, ToS, privacy policies. Full pipeline: document structure, per-section writing, Bluebook citation, case law lookup (CourtListener API), regulation lookup (eCFR API), DOCX output, and TDD-style verification…