Higgsfield Cinematic

Higgsfield Cinematic Skill

What this skill does

Crafts production-ready cinematic video prompts and routes them to the best Higgsfield model for film-grade output. On request, it assembles the full parameter set, surfaces the credit cost, and — after explicit user confirmation — generates the asset via the opt-in generation flow. The skill also handles mood-board stills via the cinematic_studio_2_5 image path.

---

Model routing

Primary and fallback video models

| | cinematicstudio30 (primary) | veo31 (fallback) | |---|---|---| | Rationale | Dedicated cinematic pipeline; default for all film-grade video | Photorealism + native audio when the studio model is unavailable or voiced output is needed | | Aspect ratios | 16:9, 9:16, 1:1 | 16:9, 9:16 only — no 1:1 | | Duration | 4–15 s (continuous range) | 4, 6, 8 s (discrete set only) | | Tunable params | None — pass only aspect_ratio and duration | quality {basic, high, ultra}; model {veo-3-1-preview, veo-3-1-fast} | | Media roles | image, start_image, end_image (optional) | start_image only (max 1) | | Native audio | No | Yes |

Routing rule: Use cinematic_studio_3_0 by default. Only fall back to veo3_1 when (1) the primary model is unavailable, or (2) the user wants native audio or explicitly photorealistic output and, after being told what they give up, accepts veo3_1's constraints: no 1:1 aspect ratio, discrete durations only (4/6/8 s instead of the continuous 4–15 s range), and start_image only (no end_image). If the user has not been informed of these trade-offs, stay on cinematic_studio_3_0 and let them decide — never switch silently.

MODEL-06 directive: If a parameter is rejected at generation time, call higgsfield:models_explore with the target model name to re-verify the live schema. The live catalog is the authoritative source — never guess parameter values. Full parameter tables: [references/model-specs.md](references/model-specs.md).

Per-platform aspect ratio and duration

Quick rule: vertical 9:16 for TikTok / Reels / Shorts, 16:9 for standard YouTube and LinkedIn, 1:1 only on cinematic_studio_3_0 (veo3_1 has none). Full per-platform recommendation table: [references/model-specs.md](references/model-specs.md).

---

Still-image path (MODEL-05)

For a production still or mood-board frame before committing to video, route to cinematic_studio_2_5 via higgsfield:generate_image. This image model supports resolution values 1k, 2k, and 4k (the 4k enum is a legitimate option for this image model only — it does not apply to any video model). Aspect ratios: 1:1, 4:3, 3:4, 16:9, 9:16. After generating the still, the output image can be passed as a start_image reference to cinematic_studio_3_0 to carry the established composition into video. Full parameter table: [references/model-specs.md](references/model-specs.md).

---

Prompt-building workflow

1. Gather intent — Confirm with the user: creative style (noir, epic,

intimate, etc.), subject, mood, target platform, aspect ratio, and desired duration.

1. Select model — Apply the routing table above. Choose

cinematic_studio_3_0 by default; select veo3_1 only when a routing trigger applies; route still requests to cinematic_studio_2_5.

1. Build the prompt — Use the craft principles in the references below to

construct the prompt:

• Opening hook (2-second attention-capture technique) from

[references/hooks.md](references/hooks.md)

• Camera movement, lighting setup, color grade, composition from

[references/camera.md](references/camera.md)

• Worked example prompts by genre/platform from

[references/examples.md](references/examples.md)

1. Present for review — Show the assembled prompt and all parameters to

the user for review and refinement before any generation call. The user must approve the prompt text, model, aspect ratio, and duration.

---

Opt-in generation

Generation costs Higgsfield credits and requires explicit user confirmation before any generate call is made. This skill never auto-generates.

Full step-by-step flow (confirmation gate, balance/cost surface, generate → poll → display): [../../shared/generation-flow.md](../../shared/generation-flow.md)

This skill's primary model: cinematic_studio_3_0

Media upload: Conditional — only when the user supplies a start_image or end_image reference for image-to-video. This skill is NOT a mandatory-I2V skill; T2V is the default path. When the user provides a reference image, run the higgsfield:media_upload → higgsfield:media_confirm atomic pair before generating.

Tool signatures: [../../shared/mcp-tools.md](../../shared/mcp-tools.md)

---

Reference materials

| File | Contents | |---|---| | [references/model-specs.md](references/model-specs.md) | Per-model parameter tables, aspect ratios, durations, routing rationale, and re-verify directive for cinematicstudio30, veo31, and cinematicstudio2_5 | | [references/camera.md](references/camera.md) | Camera movement encyclopedia (20+ moves), lighting library (15+ setups), composition rules, color-grading approaches, pacing and atmosphere guidance | | [references/hooks.md](references/hooks.md) | 2-second hook framework, 12 hook techniques with prompt phrasing templates, advanced hook-stacking guidance | | [references/examples.md](references/examples.md) | Full production-quality example prompts by genre (noir, epic landscape, dramatic dialogue, action, abstract) with timeline segmentation guides |

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: pixelab-ch
• Source: pixelab-ch/higgsfield-skills
• License: MIT

Install and usage instructions live in the source repository linked above.