video-provider-capabilities

Resolve provider capabilities before generation

Do not create a paid video request from a generic template or model-family memory. Before the first video call for a production, read the selected endpoint's current official API reference. For fal-hosted models use that exact endpoint's fal API page; for direct Google models use the current Google Gemini/Veo documentation. A model list discovers candidates; the endpoint schema controls the request. Models and higher-level workflows are different contracts. Do not infer one from the other.

Read these bundled references when their phase begins:

• references/capability-record.md before resolving or refreshing a model.
• references/continuity-planning.md before writing the shot list or request.
• references/execution-review.md before spending, retrying, assembling, or presenting outputs.
• video-model-selection/references/cost-guidance.md when comparing models or giving a user a costed choice.

Required artifacts

Write the following under the production folder in direct chat, or inside the current workflow-stage folder:

• capabilities/<provider>-<model-slug>.json: the live endpoint contract.
• continuity-plan.json: scene boundaries, reference lineage, seams, and the reason for every independent generation.
• generation-ledger.json: cost estimate, exact submitted request, provider adjustments, job state, output, validation, presentation, and review.

These are required inputs and evidence for every paid video request. The capability record begins with:

{
  "schema_version": 2,
  "provider": "fal-ai or google-ai",
  "model_id": "exact resolved model id",
  "execution_mode": "queue or realtime",
  "checked_at": "RFC3339 timestamp",
  "official_docs": ["official API URL"],
  "input_contract": {
    "required": [],
    "optional": [],
    "media_roles": {"start_frame": {"min": 0, "max": 0}, "end_frame": {"min": 0, "max": 0}, "image_references": {"min": 0, "max": 0}, "video_references": {"min": 0, "max": 0}, "audio_references": {"min": 0, "max": 0}},
    "duration_seconds": {"min": 0, "max": 0},
    "multi_shot": {"supported": false, "field": "", "shape": {}},
    "extension": {"supported": false, "input_role": ""},
    "seed": {"supported": false, "field": ""},
    "native_audio": {"supported": false, "field": "", "values": []},
    "aspect_ratios": [],
    "resolutions": [],
    "mutually_exclusive": [],
    "output_contract": {}
  },
  "request_shape": {"exact field names": "copied from the official schema"},
  "pricing": {"source": "", "checked_at": "", "currency": "", "unit": "", "unit_price": null, "base_estimate": null, "retry_reserve": null, "maximum_approved_cost": null, "estimate_available": false},
  "continuity_fit": {"strategy": "...", "why": "..."}
}