Marketing & claims legal review

When marketing copy is attached, a legal agent extracts and classifies every claim, checks substantiation and the applicable standard, and writes a claim-by-claim flagged memo with safer rewrites for an attorney to approve before publishing. Not legal advice.

Join the waitlist

What it installs

Agents 2

  • Legal Agent

    A product-legal reviewer who extracts and classifies claims, checks substantiation and the operative standard, and writes the risk-flagged memo for attorney approval.

  • Claims Verifier

    An adversarial fact-checker who confirms each cited advertising standard and substantiation claim is real and traceable before the memo proceeds to attorney review.

Teams 1

  • Claims Review Desk

    A legal reviewer who extracts and classifies claims, checks substantiation and the operative standard, and drafts the flagged memo, paired with a verifier who confirms every cited standard and substantiation call before anything reaches the attorney.

Workflows 1

  • Extract, check, research, verify, approve

    Extract and classify claims, check substantiation and the operative standard, write the flagged memo, verify the cited standards, and hand off for attorney approval.

Skills 4

  • marketing-claims-review

    Review marketing copy claim-by-claim: extract every claim, classify it on the taxonomy (vague/puffery, specific-factual, comparative, implied, absolute), check each non-puffery claim's substantiation against the specific claim (who measured it, when, sample, basis), research the operative standard for that claim, and produce a flagged memo with a 'keeps the energy' rewrite per risk. Never emits a publish approval — that is the attorney's act. Adapted from anthropics/claude-for-legal/marketing-claims-review.

  • launch-review

    Run a category-by-category legal review of a launch (contractual commitments, privacy, security, IP, third-party, regulatory, marketing claims, AI governance), calibrated to what actually blocks at this company, with sector overlays (children, fintech, health, gaming, employment) and a privileged-memo / safe-to-post-redacted dual output. Informs the approval; never grants it. Adapted from anthropics/claude-for-legal/launch-review.

  • feature-risk-assessment

    Produce a standalone deep risk assessment when a single feature or claim needs more than a line item — a novel area, a regulated sector, a regulator actively looking: 2–5 specific risks (scenario, who gets hurt, likelihood and severity with reasons, mitigations, gap, residual risk), regulatory landscape and precedent if relevant, 2–3 options, and a recommendation. Frames the decision; doesn't make it. Adapted from anthropics/claude-for-legal/feature-risk-assessment.

  • legal-risk-assessment

    Classify a legal risk on a severity × likelihood matrix (1–5 each, score 1–25) into GREEN / YELLOW / ORANGE / RED with the recommended actions, escalation criteria, and outside-counsel triggers for each band, documented in a standard risk-assessment memo. Assists the workflow; assessments are reviewed by qualified professionals. Adapted from anthropics/knowledge-work-plugins/legal-risk-assessment.

Requirements

What this template expects to do its job. Task Machine does not verify these — you decide whether your setup is ready.

  • Web search and fetch access for substantiation standards — The agent researches the currently operative advertising and substantiation standards (FTC, NAD, state UDAP, sector and platform rules) for the specific claims under review through runtime web search and fetch tools, and pauses for your approval on the memo. It reads public sources only and flags every citation for verification.

Get started

Install Marketing & claims legal review and run it with approvals.

Join the waitlist and we will send early access when the first private beta spots open.

Private beta. We invite teams in batches and never share your email.