Docs

Check continuity

Check continuity across surfaces on Stunt Double: pricing, terminology, feature names, and promises should match between marketing site, product, docs, and emails. Actors walk each surface and the mismatches surface with evidence.

When to use

  • Marketing, product, docs, and emails may have drifted out of sync (pricing, plan names, feature terminology, promises)
  • A pricing or naming change shipped on one surface but maybe not the others
  • You want cross-surface drift caught as content ships, not by customers

Instructions

  1. Resolve context:

    • list_workspaces() → pick the workspace
    • list_projects(workspace_id) → match the surfaces to a project, or let create_checklist find/create it
  2. Ensure a reviewer actor exists:

    • list_actors(workspace_id), else create_actor(...)
  3. Create a checklist per surface with mirrored extraction checks:

    • Record the same facts on each surface so they can be compared:
      • "Record the price, billing period, and plan names shown"
      • "Record what the main feature is called on this page"
    • Plus cross-surface assertions on the primary surface:
      • "The price at checkout matches the price on the pricing page"
      • "The feature is named consistently; it is never called by an old name"
  4. For promise-versus-product continuity, use an interview:

    • create_interview(...) with a task that walks from the marketing claim to using the feature ("The homepage says setup takes two minutes; sign up and see if that holds"), a persona participant, and launch_interview
    • Poll and read get_interview_report
  5. Run everything and compare recorded values across surfaces:

    • run_checklist per surface → poll get_checklist_run until terminal
  6. Report a discrepancy table:

    • Each item, what every surface shows, which surfaces disagree, with evidence
    • Rank by customer impact (billing mismatches first)
  7. Offer standing coverage:

    • A scheduled create_workflow so drift is caught as content ships (see setup-guardrails)

Example output

Continuity check: marketing site vs app vs docs

| Item              | Marketing        | App (checkout)   | Docs             | Disagree?     |
| ----------------- | ---------------- | ---------------- | ---------------- | ------------- |
| Pro price / month | $29              | $39              | $29              | App ≠ others  |
| Plan names        | Starter/Pro/Team | Starter/Pro/Team | Starter/Pro/Biz  | Docs: "Biz"   |
| Core feature name | "Flows"          | "Flows"          | "Workflows"      | Docs: old name|
| Setup promise     | "2-minute setup" | ~6 min in interview transcript |    | Promise > reality |

Ranked: billing mismatch first (App charges $39 vs $29 advertised). Then docs plan-name + feature-name drift.
Offer: schedule this so drift is caught as content ships.

Tips

  • Mirror the extraction checks across surfaces so the recorded values line up cleanly for comparison.
  • Billing mismatches rank first: a price that disagrees between the pricing page and checkout is the highest-impact discrepancy.
  • Interviews beat checklists for promises. A task that walks from the marketing claim to the real experience captures whether the promise holds.

Edit on GitHub

On this page