brainstorming
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or planning anything new, in software or out of it (a talk, a business, a renovation).
runway-hello-worldCreate, observe, and preserve one approved Runway text-to-video task using the current per-model contract. Use when proving the first billable generation path. Trigger with: "Runway hello world", "first Runway video", "smoke test Runway generation".
Independent GitHub snapshot · Automated classification · Original documentation governs
name: runway-hello-world description: >- Create, observe, and preserve one approved Runway text-to-video task using the current per-model contract. Use when proving the first billable generation path. Trigger with: "Runway hello world", "first Runway video", "smoke test Runway generation". allowed-tools: Read, Grep, Write, Edit version: 2.0.0 argument-hint: '[prompt-and-model-policy]' model: inherit effort: high license: MIT author: Jeremy Longshore jeremy@intentsolutions.io tags:
Prove the minimum asynchronous generation path without pretending that task creation is completion. A Runway generation consumes credits and may be moderated, so the smoke test requires an explicit cost ceiling, safe prompt, terminal-state evidence, and owned output storage.
Open the current models guide and the exact text_to_video model variant in the API reference. Confirm model identifier, required fields, prompt limit, ratio set, duration range, and price immediately before the test.
Record a unique operation ID, normalized request fingerprint, model, duration, ratio, estimated ceiling, and approval. Use gen4.5 only when its current schema still matches the request; never revive retired gen3a_turbo.
Submit one server-side request. Persist the returned task ID before waiting. Treat the create response as acceptance into the task system, not a generated asset.
Use the SDK wait helper or poll GET /v1/tasks/<task-id> no more frequently than every five seconds with jitter and backoff. Set a deadline. Remember that a client timeout does not cancel the provider task.
Accept output only from SUCCEEDED. For FAILED, retain redacted failureCode and diagnostic text; for CANCELLED, record who cancelled. Treat THROTTLED as queued, not failed.
Download successful output promptly, validate type and size, store it under the operation ID, and retain the source URL only as sensitive temporary evidence because Runway output URLs expire.
Create and task-read requests use the server-side bearer secret and reviewed Runway API version. Never expose the key or temporary signed output URL to an untrusted browser, log stream, or ticket.
Use Read and Grep to inspect application configuration, provider documentation, lockfiles, fixtures, schemas, tests, and redacted operational evidence before proposing a change. Use Write or Edit only for an approved implementation, configuration, test, runbook, or redacted receipt. Do not create, cancel, delete, retry, deploy, rotate, revoke, publish, or otherwise mutate production Runway resources without explicit operator approval.
Return the environment, organization alias, operation and task identifiers, API and SDK versions, model or router policy, source-contract fingerprint, task-state evidence, credit boundary, output disposition, unresolved risk, rollback state, and final decision without exposing API secrets, prompt or media contents, or temporary signed URLs.
An operator approves one five-second gen4.5 test under a fixed credit ceiling. The worker persists the task ID, observes PENDING, RUNNING, and SUCCEEDED, copies the video into private storage, and records its checksum without publishing the temporary provider URL.
| Failure | Response |
|---|---|
Task stays THROTTLED | Keep the task queued, observe concurrency and daily limits, and do not submit duplicates. |
| Wait helper times out | Retrieve the saved task ID later or explicitly cancel it; timeout did not cancel the task. |
| Task fails moderation | Do not retry the same prompt; record the safety code without exposing unsafe content and review policy. |
Replay the workflow with a safe approved canary, prove task ID persistence before polling, exercise one mocked failure and timeout path, and confirm the stored output remains available after discarding the temporary URL.