ThomasMoreAI/legal-skills-open / au/general/skills/route-legal-fact-pattern-l0cka
route-legal-fact-pattern-l0cka
VerifiedRoute a described Australian legal matter or fact pattern to the Legal Skills plugins and skills it engages and return a sequenced routing plan — configuration and profile skills first, official-source verification next, assessment and mapping skills after, deliverable skills last — with the human decision owner at each hand-off and the plugins still to install, using the generated skill map. Use first when a matter spans more than one practice area (a data breach at a listed company, an AI deployment by a regulated entity, a deceased estate with a limitation question) or when the user is unsure which skill to run. Do not use to answer the legal question, assess merits, run the routed skills, or route a practice area no shipped skill covers.
Route Legal Fact Pattern
Turn a described matter into an ordered list of the skills to run, and say what stays with a human at each step. The router is a map, not an adviser: every routed skill keeps its own scope, fail-closed rules and review points.
Read references/skill-map.md before routing. It is generated from the marketplace registry and is the only list of skills the router may name; never route to a skill that is not in it. The map lists every shipped skill, not the skills installed in this session: each plugin installs separately, and the map records which plugins a plugin requires or hands off to.
Workflow
- Extract the routing facts.
- Record, with the user's wording: parties and their roles (entity type, listed or regulated status, individual or organisation); sector and regulator overlays; jurisdiction or jurisdictions; the events and their dates; existing profiles, registers or approved materials the user already holds; and the deliverable sought.
- Record
cannot be determinedfor any routing fact the description does not supply. Ask one bounded question only where the answer changes the plan; otherwise route on stated assumptions and label them.
- Match facts to skills.
- For each fact, list every skill in the map whose description is triggered by it and note which of its stated exclusions might apply.
- Prefer the most specific skill. Where an orchestrating skill exists (for example a breach assessment that itself calls the APP checker and the legislation verifiers), route to the orchestrator and list its dependencies as prerequisites rather than as separate steps.
- Where a fact engages a practice area with no skill in the map, record a gap. Never route it to the nearest skill.
- Add each routed plugin's required plugins from the map as prerequisites. Add a hand-off plugin only when a fact engages it.
- Check availability.
- For each routed skill, record
installedwhen it appears in this session's available skills,not installedwhen the session's skill list is visible and it is absent, andcannot be determinedwhen the session does not expose its skill list. - Collect every
not installedplugin, required prerequisites included, under "Install first". A missing plugin is not a gap: keep its step in the plan and never replace it with an installed skill that looks close.
- For each routed skill, record
- Sequence the plan.
- Order: (a) configuration and profile skills the routed skills require; (b) official-source verification of the legislation and case law the later steps rely on; (c) assessment, mapping and computation skills; (d) deliverable and record skills (registers, chronologies, drafts, board papers).
- Mark each step with its inputs (including outputs of earlier steps), the plugin it belongs to, and the human decision owner who must act on its output before the next step runs.
- Mark any step whose skill fails closed on a missing profile or unverified table, so the user runs the prerequisite first.
- Report using the result contract.
Result contract
Lead with READY FOR HUMAN REVIEW, READY WITH QUALIFICATIONS, NOT READY
or OUTSIDE SCOPE.
Matter: <label; jurisdiction(s); parties and roles; deliverable sought>
Routing facts: <fact; provenance in the description; assumption flag>
Plan:
<step>. <plugin> / <skill> — <why engaged>; inputs: <...>; availability:
<installed | not installed | cannot be determined>; human decision:
<owner and decision before the next step>
Install first: <plugins not installed, with the steps that need them, or none>
Gaps: <fact patterns or practice areas no shipped skill covers>
Assumptions and questions: <stated assumptions; one bounded question if any>
Limitations: <the plan is a map, not advice; each skill's own scope governs>
Use READY FOR HUMAN REVIEW when every routing fact has provenance, every
step names a skill from the map and every routed skill is installed. Use
READY WITH QUALIFICATIONS when the plan rests on labelled assumptions,
contains gaps, or has a step whose skill is not installed or whose
availability cannot be determined. Use NOT READY when the
description does not identify a jurisdiction, a party role or a deliverable
and no reasonable assumption can be stated. Use OUTSIDE SCOPE when the
request is for the legal answer, a merits view, or a jurisdiction outside
Australia.
Fail closed
Never answer the legal question, assess merits, or describe the plan as advice. Never name a skill absent from the skill map, invent a plugin, or route an uncovered practice area to the nearest skill. Never run a routed skill from inside the router or represent a routed skill's output. Never drop a human decision owner from a step to make the plan shorter. Never report a skill as installed without seeing it in this session's available skills, and never drop a step or its required prerequisites because the plugin is not installed.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| writing-skillsobra/superpowersUse when creating new skills, editing existing skills, or verifying skills work before deployment | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| xlsxanthropics/skillsUse this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .xltx, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like "the xlsx in my downloads") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| shipping-and-launchaddyosmani/agent-skillsPrepares production launches. Use when preparing to deploy to production, or when asking what needs to be in place before shipping. Use when you need a pre-launch checklist, when setting up monitoring, when planning a staged rollout, or when you need a rollback strategy. | Stars: 119k | Updated | Claude CodeCodex | Verified | Install |
| childhoodthedotmack/claude-memBabysit a shipped release after merge, publication, or deployment. Verify published packages, production behavior, installed plugins, and late CI or user reports until the observation window is clean; use babysit for PR readiness before release. | Stars: 118k | Updated | Claude CodeCodex | Verified | Install |