legal-writing
Skill1 starsChanged 8 days agoMITView on GitHub
Use for Australian legal drafting: affidavits, submissions, orders, statutory declarations, deeds, demand and pre-action letters, decision overviews, AGLC4 citation, and condensing any of them. Not for general or academic prose about law (natural-writing) or engineering docs (engineering-writing). Drafting assistance, not legal advice.
2 other projects use a version of this.
Adopted by
Most recently taken up by
- catalsqlb-cmd/Prism-Htheir own version, first seen May 29, 2026
- anthropics/claude-for-legaltheir own version, first seen May 15, 2026
Other versions2 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 2
- anthropics/claude-for-legal★ 9.5ksame name
- catalsqlb-cmd/Prism-H★ 16same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository21 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- code-reviewSkill
- deliverSkill
- diagnoseSkill
- engineering-docsSkill
- engineering-writingSkill
- evaluateSkill
- implementSkill
- natural-writingSkill
- orchestrateSkill
- refactorSkill
- releaseSkill
- retrospectSkill
- scopeSkill
- sessionSkill
- setup-repoSkill
- skill-craftSkill
- test-designSkill
- trackerSkill
- ui-ux-designSkill
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 5 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 19
Use it when the output is an Australian legal instrument, filing-facing text or legal correspondence: court and tribunal forms, affidavits, …
- Distinctive: line 32
Read project instructions, live matter state and source-boundary rules.
- Distinctive: line 46
Run the lint, then the checks the task needs.
- Distinctive: line 58
Every filing-facing fact needs a verified real source anchor.
- Distinctive: line 62
Never humanise by weakening a threshold, deleting an anchor, casualising court language or changing forum wording.
- Very specific: line 67
Use research for current law.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Australian legal writing
Stricter project and forum instructions prevail. Verify current law, forms and procedure against official sources, and keep qualified human review for filing-facing work. Sending, signing, filing and serving stay with the user.
The natural-writing hub owns Australian English, the mechanism catalogue and
the condense pass. This skill owns jurisdiction, forum, instrument and
filing-facing rules. Its Legal Function Test classifies sentences by legal
function and does not map to the hub's observed/inferred claim classes.
Scope
Use it when the output is an Australian legal instrument, filing-facing text or legal correspondence: court and tribunal forms, affidavits, statutory declarations, witness statements, submissions, proposed orders, chronologies, annexure indexes, deeds and agreements, letters of demand, registry and party correspondence, non-filing decision overviews and AGLC4 citation.
Legal subject matter alone is not the trigger. An article, chapter or
plain-language explanation about law goes to natural-writing; a spec, README or
runbook goes to engineering-writing. This skill joins either as a companion
where legal status, authority, forum wording or source altitude must survive.
Workflow
- Read project instructions, live matter state and source-boundary rules. Load the matching jurisdiction, forum or document skill where one exists.
- Choose the mode: draft, rewrite, condense, diagnose, correspondence, decision-overview or final-scrub. Load legal concision for all prose. validation-checklists.md maps the other references to mode and document type.
- Classify each paragraph by legal function before polishing. Move wrong-home material first.
- Lock legal status, source anchors, exact quotations, authorities, labels, offer terms, non-admission and non-waiver wording, deadlines, attachments, redactions and user-authority conditions.
- Draft the minimum complete text: relief, request or answer first, one proposition per paragraph, exact pinpoints, no internal agent language.
- Run the lint, then the checks the task needs.
provenant check writing path/to/source --rules legalThe lint is a guardrail. Source checks, forum skills, render checks and user gates still apply. Edits to filing-facing, source-boundary or lint rules need independent review.
Bounds
- Every filing-facing fact needs a verified real source anchor. Never invent or silently alter an authority, instrument, finding, rule title, forum term or decision-maker label. Verify canonical sources, not indexes, OCR, renders or agent notes.
- Never humanise by weakening a threshold, deleting an anchor, casualising court language or changing forum wording.
- Preserve contentions as contentions, in the register the stage requires. Protective-order, police-issued or safety material stays a dated source with the legal effect its verified text supports, not a finding.
- Use
researchfor current law; claims cite primary sources and retrieval dates. - Condensing runs the token set-diff plus an independent qualitative pass. Final scrub adds no argument, fact, authority, history or courtesy closer.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

