Skip to content
Get packs

AI Use Case Triage

Classify a proposed AI use case as approved, conditional, or not approved, with required conditions and next steps.

Source Download skill
Repository
anthropics/claude-for-legal
Stars
9,870
Licence
Apache-2.0
Files
1 file

What it does

Once installed, the agent checks a proposed use of AI (such as using AI to score sales leads) against your organisation's written registry of approved AI uses and red lines, then labels it approved, conditional, or not approved. For conditional cases it lists the concrete conditions and the next steps, and it flags where privacy or product counsel needs to be involved. If a use case is not in the registry, it defaults to conditional pending a formal AI impact assessment rather than guessing.

Great for

  • In-house legal or compliance teams who field 'can we use AI for this?' requests from the business
  • Governance leads who want a consistent first-pass triage before a full impact assessment
  • Teams that have already populated their AI use case registry and red lines

Skip it if

  • Getting a binding legal opinion: it gives a calibrated first pass against the configured registry, and the output still needs lawyer review (the skill itself tags regulatory citations as needing verification).
  • Organisations without a configured registry: it can run in 'provisional' mode against generic defaults, which may not reflect your actual decisions or jurisdiction.
  • Jurisdictions outside the configured regulatory footprint: the skill says it will flag these for re-triage rather than extend by analogy, so it is unreliable for them until configured.

Try asking

  • “Triage this use case: our sales team wants to score leads with AI automatically”
  • “Can we use AI to summarise customer support tickets? Is this approved?”
  • “What do we need to do to use AI for drafting job descriptions?”
  • “batch: triage this list of five proposed AI tools from the operations team”
Step-by-step guide: how to use AI Use Case TriageInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 1 original files, the Apache-2.0 licence and an attribution note.

    use-case-triage.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip use-case-triage.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip use-case-triage.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Run the cold-start interview first so the registry, jurisdiction and risk appetite are your own rather than generic defaults.
  • Describe the use case concretely: what the AI does, who it acts on, whether a human reviews output, and which vendor is involved.
  • Treat the classification as a first pass and check any regulatory citations flagged for verification before relying on them.

What's inside

  • SKILL.md

Redistributed unmodified from anthropics/claude-for-legal under the Apache-2.0 licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.