{"id":"eb4ca4e6-cef7-4e5e-afd0-7464085a91f1","entityType":"agent","slug":"clawhub-higgsfield-higgsfield-generate","name":"Higgsfield Generate","canonicalUrl":"https://www.xpersona.co/agent/clawhub-higgsfield-higgsfield-generate","canonicalPath":"/agent/clawhub-higgsfield-higgsfield-generate","generatedAt":"2026-10-09T22:37:11.243Z","source":"CLAWHUB","claimStatus":"UNCLAIMED","verificationTier":"NONE","summary":{"evidence":{"source":"editorial-content","verified":true,"confidence":"high","updatedAt":"2026-10-09T17:07:33.531Z","emptyReason":null},"description":"Generate images and videos via Higgsfield AI through 30+ models including Nano Banana 2, Soul V2, Veo 3.1, Kling 3.0, Seedance 2.0, Flux 2, GPT Image 2, plus... Skill: Higgsfield Generate Owner: higgsfield Summary: Generate images and videos via Higgsfield AI through 30+ models including Nano Banana 2, Soul V2, Veo 3.1, Kling 3.0, Seedance 2.0, Flux 2, GPT Image 2, plus... Tags: latest:1.0.0 Version history: v1.0.0 | 2026-05-04T16:20:37.784Z | user - Initial release of higgsfield-generate for image and video generation via Higgsfield AI, supporting 30+ models. - Supports a w","descriptionLabel":"Technical summary","evidenceSummary":"Capability contract not published. No trust telemetry is available yet. 2.2K downloads reported by the source. Last updated 10/9/2026.","installCommand":"clawhub skill install s17f8fwnpnszqskag8vjqhx1ts862976:higgsfield-generate","sourceUrl":"https://clawhub.ai/higgsfield/higgsfield-generate","homepage":"https://clawhub.ai/higgsfield/skills/higgsfield-generate","primaryLinks":[{"label":"View on ClawHub","url":"https://clawhub.ai/higgsfield/higgsfield-generate","kind":"source"},{"label":"Homepage","url":"https://clawhub.ai/higgsfield/skills/higgsfield-generate","kind":"homepage"}],"safetyScore":84,"overallRank":62,"popularityScore":67,"trustScore":null,"claimedByName":null,"isOwner":false,"seoDescription":"Generate images and videos via Higgsfield AI through 30+ models including Nano Banana 2, Soul V2, Veo 3.1, Kling 3.0, Seedance 2.0, Flux 2, GPT Image 2, plus..."},"coverage":{"evidence":{"source":"public-profile","verified":false,"confidence":"medium","updatedAt":"2026-10-09T17:07:33.531Z","emptyReason":null},"protocols":[{"protocol":"OPENCLEW","label":"OpenClaw","status":"self-declared","notes":"Declared in the public agent profile."}],"capabilities":[],"verifiedCount":0,"selfDeclaredCount":1,"capabilityMatrix":{"rows":[{"key":"OPENCLEW","type":"protocol","support":"unknown","confidenceSource":"profile","notes":"Listed on profile"}],"flattenedTokens":"protocol:OPENCLEW|unknown|profile"}},"adoption":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-09T17:07:33.531Z","emptyReason":null},"stars":null,"forks":null,"downloads":2244,"packageName":null,"latestVersion":"1.0.0","tractionLabel":"2.2K downloads"},"release":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-09T17:07:33.531Z","emptyReason":null},"lastUpdatedAt":"2026-10-09T17:07:33.531Z","lastCrawledAt":"2026-10-09T17:07:33.531Z","lastIndexedAt":null,"nextCrawlAt":"2026-10-10T17:07:33.531Z","lastVerifiedAt":null,"highlights":[{"version":"1.0.0","createdAt":"2026-05-04T16:20:37.784Z","changelog":"- Initial release of higgsfield-generate for image and video generation via Higgsfield AI, supporting 30+ models. - Supports a wide range of tasks: text-to-image, image-to-image, image-to-video, reference-based generation, and branded ad creation through Marketing Studio. - Automatic model selection and input handling with concise, user-friendly interaction rules. - Includes dedicated workflows and parameter flags for both generic generation and branded ad content (avatars, products, marketing videos/images). - Ensures easy CLI installation, authentication, and result retrieval, delivering output URLs with brief summaries.","fileCount":10,"zipByteSize":18009}]},"execution":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No published capability contract is available yet."},"installCommand":"clawhub skill install s17f8fwnpnszqskag8vjqhx1ts862976:higgsfield-generate","setupComplexity":"low","setupSteps":["Setup complexity is LOW. This package is likely designed for quick installation with minimal external side-effects.","Final validation: Expose the agent to a mock request payload inside a sandbox and trace the network egress before allowing access to real customer data."],"contract":{"contractStatus":"missing","authModes":[],"requires":[],"forbidden":[],"supportsMcp":false,"supportsA2a":false,"supportsStreaming":false,"inputSchemaRef":null,"outputSchemaRef":null,"dataRegion":null,"contractUpdatedAt":null,"sourceUpdatedAt":null,"freshnessSeconds":null},"invocationGuide":{"preferredApi":{"snapshotUrl":"https://www.xpersona.co/api/v1/agents/clawhub-higgsfield-higgsfield-generate/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-higgsfield-higgsfield-generate/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-higgsfield-higgsfield-generate/trust"},"curlExamples":["curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-higgsfield-higgsfield-generate/snapshot\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-higgsfield-higgsfield-generate/contract\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-higgsfield-higgsfield-generate/trust\""],"jsonRequestTemplate":{"query":"summarize this repo","constraints":{"maxLatencyMs":2000,"protocolPreference":["OPENCLEW"]}},"jsonResponseTemplate":{"ok":true,"result":{"summary":"...","confidence":0.9},"meta":{"source":"CLAWHUB","generatedAt":"2026-10-09T22:37:11.242Z"}},"retryPolicy":{"maxAttempts":3,"backoffMs":[500,1500,3500],"retryableConditions":["HTTP_429","HTTP_503","NETWORK_TIMEOUT"]}},"endpoints":{"dossierUrl":"https://www.xpersona.co/api/v1/agents/clawhub-higgsfield-higgsfield-generate/dossier","snapshotUrl":"https://www.xpersona.co/api/v1/agents/clawhub-higgsfield-higgsfield-generate/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-higgsfield-higgsfield-generate/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-higgsfield-higgsfield-generate/trust"}},"reliability":{"evidence":{"source":"runtime-metrics","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No trust, reliability, or runtime telemetry is available."},"trust":{"status":"unavailable","handshakeStatus":"UNKNOWN","verificationFreshnessHours":null,"reputationScore":null,"p95LatencyMs":null,"successRate30d":null,"fallbackRate":null,"attempts30d":null,"trustUpdatedAt":null,"trustConfidence":"unknown","sourceUpdatedAt":null,"freshnessSeconds":null},"decisionGuardrails":{"doNotUseIf":["Contract metadata is missing or unavailable for deterministic execution."],"safeUseWhen":[],"riskFlags":["missing_or_unavailable_contract","trust_data_unavailable","schema_references_missing"],"operationalConfidence":"low"},"executionMetrics":{"observedLatencyMsP50":null,"observedLatencyMsP95":null,"estimatedCostUsd":null,"uptime30d":null,"rateLimitRpm":null,"rateLimitBurst":null,"lastVerifiedAt":null,"verificationSource":null},"runtimeMetrics":{"successRate":null,"avgLatencyMs":null,"avgCostUsd":null,"hallucinationRate":null,"retryRate":null,"disputeRate":null,"p50Latency":null,"p95Latency":null,"lastUpdated":null}},"benchmarks":{"evidence":{"source":"no-benchmark-data","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No benchmark suites or observed failure patterns are available."},"suites":[],"failurePatterns":[]},"artifacts":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"high","updatedAt":"2026-10-09T17:07:33.531Z","emptyReason":null},"readme":"Skill: Higgsfield Generate\n\nOwner: higgsfield\n\nSummary: Generate images and videos via Higgsfield AI through 30+ models including Nano Banana 2, Soul V2, Veo 3.1, Kling 3.0, Seedance 2.0, Flux 2, GPT Image 2, plus...\n\nTags: latest:1.0.0\n\nVersion history:\n\nv1.0.0 | 2026-05-04T16:20:37.784Z | user\n\n- Initial release of higgsfield-generate for image and video generation via Higgsfield AI, supporting 30+ models.\n- Supports a wide range of tasks: text-to-image, image-to-image, image-to-video, reference-based generation, and branded ad creation through Marketing Studio.\n- Automatic model selection and input handling with concise, user-friendly interaction rules.\n- Includes dedicated workflows and parameter flags for both generic generation and branded ad content (avatars, products, marketing videos/images).\n- Ensures easy CLI installation, authentication, and result retrieval, delivering output URLs with brief summaries.\n\nArchive index:\n\nArchive v1.0.0: 10 files, 18009 bytes\n\nFiles: references/marketing-avatars.md (1080b), references/marketing-modes.md (2966b), references/marketing-products.md (1456b), references/media-inputs.md (3901b), references/model-catalog.md (11529b), references/prompt-engineering.md (1554b), references/troubleshooting.md (1196b), skill-card.md (2624b), SKILL.md (11384b), _meta.json (138b)\n\nFile v1.0.0:SKILL.md\n\n---\nversion: 0.3.0\nname: higgsfield-generate\ndescription: |\n  Generate images and videos via Higgsfield AI through 30+ models including\n  Nano Banana 2, Soul V2, Veo 3.1, Kling 3.0, Seedance 2.0, Flux 2, GPT Image 2,\n  plus Marketing Studio for branded ad video/image with curated avatars and\n  imported products.\n  Use when: \"generate an image\", \"make a picture\", \"create artwork\",\n  \"make a video\", \"animate this photo\", \"image-to-video\", \"img2vid\",\n  \"edit this image with AI\", \"stylize a photo\", \"remix this image\",\n  \"produce a clip\", \"render a scene\", \"create an ad\", \"make a UGC video\",\n  \"generate marketing video\", \"make a product demo\", \"create unboxing\",\n  \"TV spot\", \"virtual try-on\", \"product showcase\", \"brand video\",\n  \"presenter video for product\", \"import product from URL\",\n  \"create avatar for ad\".\n  Supports text-to-image, image-to-image, image-to-video, reference-based\n  generation, and Marketing Studio (avatars + products + ad modes).\n  Auto-detects whether passed IDs are uploads or previous jobs.\n  Chain with higgsfield-soul-id when the user wants their face in the output.\n  NOT for: training Soul Character (use higgsfield-soul-id), professional product\n  photoshoots with mode-specific prompt enhancement (use\n  higgsfield-product-photoshoot), text-only / chat / TTS tasks.\nargument-hint: \"[prompt] [--model <name>] [--image <path-or-id>]\"\nallowed-tools: Bash\n---\n\n# Higgsfield Generate\n\nSubmit jobs to any Higgsfield model. Wraps the `higgsfield` CLI. Covers generic image/video gen and Marketing Studio (branded ads, avatars, products).\n\n## Step 0 — Bootstrap\n\nBefore any other command, make sure the CLI is installed and authenticated:\n\n1. If `higgsfield` is not on `$PATH`, install it:\n   ```bash\n   curl -fsSL https://raw.githubusercontent.com/higgsfield-ai/cli/main/install.sh | sh\n   ```\n2. If `higgsfield account status` fails with `Session expired` / `Not authenticated`, ask the user to run `higgsfield auth login` (interactive, opens a browser) and wait for them to confirm before continuing.\n\nSkip both checks if `higgsfield account status` already prints account info.\n\n## UX Rules\n\n1. Be concise. No raw IDs, no JSON dumps in chat. Print result URL when ready.\n2. No internal jargon. Don't narrate \"calling higgsfield cost\", \"polling job\".\n3. Detect the user's language from the first message and reply in it. Technical args (`--aspect_ratio 16:9`) stay English.\n4. Don't batch-ask. Pick a sane default model and ask one thing at a time only if genuinely missing.\n5. Don't pre-estimate cost. Just submit unless the user asks.\n6. Pass `--wait` to `generate create` so the command blocks until done and prints the result URL itself. Avoid the two-step `create` → `wait` pattern.\n\n## Workflow — generic generation\n\n1. **Pick a model.** Practical defaults from production use:\n\n   **Image:**\n   - Brand product visual (Pinterest pin, lifestyle, hero banner, ad pack, virtual try-on) → use `higgsfield-product-photoshoot` instead. NOT this skill.\n   - Branded ad image with avatar + product (Marketing Studio shape) → Marketing Studio Image (see Marketing Studio below)\n   - Aesthetic UGC / fashion editorial / lifestyle character → Soul 2.0\n   - Cinematic still frame → Soul Cinema\n   - Highly characterful creative persona (text-only, distinctive) → Soul Cast\n   - Locations / environments / no-people scenes → Soul Location (best in class)\n   - Vector illustrations OR face edit + complex scene swap → Seedream 4.5\n   - Soul Character (reference id from `higgsfield-soul-id`) → Soul 2.0 for stills, Soul Cinema for cinematic\n   - Fast and cheap iteration → Z Image\n   - Character or cartoon-style work → Nano Banana 2; step up to Nano Banana Pro on hard cases\n   - **Default for everything else → GPT Image 2.** Graphic design, UI, banners, typography, and high-fidelity general generation.\n\n   **Video:**\n   - All advertising / commercial / branded ad video → Marketing Studio (see Marketing Studio below)\n   - **Default all-purpose serious video (multi-shot, consistent identity, motion-heavy) → Seedance 2.0.** SOTA.\n   - Single-plane scene without strong dynamics, cheaper than Seedance 2.0 → Kling 3.0\n   - Cheap clean shot without cuts → Seedance 1.5 Pro\n   - Cinema-grade highest fidelity → Cinema Studio Video 3.0\n   - Cheap with strong physics, no audio needed → Minimax Hailuo\n   - Fast batch / volume → Veo 3.1 Lite\n\n   For the actual `--model` ID to pass to `higgsfield generate create`, run `higgsfield model list --json | jq` to map display names to IDs. See `references/model-catalog.md` for the full table.\n\n2. **Pass media inputs straight to flags.** Media flags accept a local file path **or** a UUID. CLI auto-uploads paths and auto-detects job vs upload for UUIDs. No need to pre-upload. Each model declares accepted roles (`image`, `start_image`, `end_image`, `video`, `audio`) — see `references/media-inputs.md`.\n3. **Validate quickly.** If unsure of params, run `higgsfield model get <jst> --json` once and pass only what's needed. Use schema defaults otherwise. The server returns `adjustments` for non-fatal coercions (e.g. `aspect_ratio=99:99` → closest match) and a structured error for invalid declared-param values.\n4. **Submit and wait in one shot.** `higgsfield generate create <jst> --prompt \"...\" [media flags] [param flags] --wait`. Blocks until terminal status and prints the result URL on stdout. Tunables: `--wait-timeout 20m` (default 10m), `--wait-interval 5s` (default 3s).\n5. **Deliver.** Send the URL plus a one-line summary (model, duration if video).\n\nTo inspect or rerun later, `higgsfield generate list --json` and `higgsfield generate get <id> --json` work for retrospection. `higgsfield generate wait <id>` is still available if you ever need to rejoin a job started without `--wait`.\n\n## Media flags\n\n| Flag | Use for | Models that accept it |\n|---|---|---|\n| `--image <path-or-id>` | reference image | most image models, `seedance_2_0`, `veo3`, `marketing_studio_video` |\n| `--start-image <path-or-id>` | first frame for image-to-video transitions | `kling3_0`, `kling2_6`, `veo3_1`, `seedance_2_0`, `marketing_studio_video` |\n| `--end-image <path-or-id>` | last frame for transitions | `kling3_0`, `seedance_2_0`, `marketing_studio_video` |\n| `--video <path-or-id>` | reference video | `seedance_2_0` |\n| `--audio <path-or-id>` | reference audio (lipsync, soundtrack match) | `seedance_2_0` (use this, NOT `--generate-audio`) |\n\nEach flag accepts either a local file path (auto-uploaded) or a UUID (upload id from `higgsfield upload create`, or a previous job id). Each model declares its own role set via `MEDIA_ROLES`. See `references/media-inputs.md` for the full table.\n\n## Common params\n\nFlags pass through to model schema. Use `higgsfield model get <jst>` to discover.\n\n```bash\nhiggsfield generate create gpt_image_2 --prompt \"neon city at dusk\" --aspect_ratio 16:9 --resolution 2k --wait\nhiggsfield generate create nano_banana_2 --prompt \"anime character concept, expressive pose\" --image ./ref.png --wait\nhiggsfield generate create seedance_2_0 --prompt \"camera dollies in\" --start-image ./first.png --duration 8 --wait\nhiggsfield generate create text2image_soul_v2 --prompt \"...\" --soul-id <soul_ref_id> --wait\n```\n\nFor machine-readable output (chained pipelines, agent context), add `--json`. With `--wait --json` you get the final job object array. Without `--wait`, you get the job IDs.\n\nStdin prompt: `echo \"...\" | higgsfield generate create z_image --wait`.\n\n## Marketing Studio\n\nBranded image/video gen: avatars + products + ad-style modes. Use models `marketing_studio_video` and `marketing_studio_image`.\n\n### Concepts\n\n- **Avatar** — presenter face. Curated `preset` (browse `higgsfield marketing-studio avatars list`) or `custom` (uploaded photos via `higgsfield marketing-studio avatars create`).\n- **Product** — brand item with title + reference images. Imported from URL (`higgsfield marketing-studio products fetch --url ...`) or created from uploaded images (`higgsfield marketing-studio products create`).\n- **Webproduct** — App Store / web page version. Auto-routes when fetching App Store URLs.\n\n### UX rules (additional)\n\n- One question per phase. Don't ask product+avatar+mode upfront.\n\n### Workflow — quick ad video\n\n1. **Get product.**\n   - URL → `higgsfield marketing-studio products fetch --url <url> --wait` (polls until import done)\n   - Local images → `higgsfield upload create <photo>...` then `higgsfield marketing-studio products create --title \"...\" --image <id>...`\n   Capture product id.\n2. **Pick avatar.**\n   - Default: `higgsfield marketing-studio avatars list` and pick a preset matching the brand voice.\n   - Custom: `higgsfield marketing-studio avatars create --name \"...\" --image <upload_id>`.\n3. **Pick mode.** Default `ugc`. Other slugs (canonical from MCP): `ugc_how_to`, `ugc_unboxing`, `product_showcase`, `product_review`, `tv_spot`, `wild_card`, `ugc_virtual_try_on`, `virtual_try_on`. See `references/marketing-modes.md`.\n4. **Generate (one-shot).**\n   ```bash\n   higgsfield generate create marketing_studio_video \\\n     --prompt \"...\" \\\n     --avatars '[{\"id\":\"<avatar_id>\",\"type\":\"preset\"}]' \\\n     --product_ids '[<product_id>]' \\\n     --mode ugc \\\n     --duration 15 \\\n     --resolution 720p \\\n     --aspect_ratio 9:16 \\\n     --wait\n   ```\n   Resolution is `480p` or `720p`. Aspect ratio is one of `auto`/`21:9`/`16:9`/`4:3`/`1:1`/`3:4`/`9:16`. `--generate-audio true` is supported here (unlike `seedance_2_0`). `--wait` blocks until done; bump `--wait-timeout 30m` for longer ad runs.\n5. **Deliver.** URL + one-line summary (mode, duration).\n\n### Click-to-Ad shortcut (URL-driven)\n\nWhen the user gives a product URL and wants a marketing video in one go:\n\n```bash\n# 1. Trigger fetch (returns the product id and starts background scrape)\nhiggsfield marketing-studio products fetch --url https://shop.example.com/sneakers --wait\n\n# 2. Generate the marketing video against the same URL — backend reuses the entity\nhiggsfield generate create marketing_studio_video \\\n  --url https://shop.example.com/sneakers \\\n  --mode ugc \\\n  --duration 15 \\\n  --aspect_ratio 9:16 \\\n  --wait\n```\n\nBackend dedupes by URL, so repeated runs reuse the existing entity instead of re-fetching.\n\n### Workflow — marketing image\n\nSame as above but use `marketing_studio_image` model:\n\n```bash\nhiggsfield generate create marketing_studio_image \\\n  --prompt \"...\" \\\n  --aspect_ratio 1:1 \\\n  --resolution 2k \\\n  --wait\n```\n\n## Errors\n\n- `Missing required params: prompt` → user gave no prompt; ask for it.\n- `Invalid values: aspect_ratio=99:99 (allowed: ...)` → bad enum; pick from allowed.\n- `Unknown params: foo` → schema doesn't accept that flag; check `higgsfield model get <jst>`.\n- `Session expired` → `higgsfield auth login`.\n\nSee `references/troubleshooting.md` for more.\n\n## Reference docs\n\nLoad on demand:\n\n- `references/model-catalog.md` — picking the right model for the task\n- `references/prompt-engineering.md` — writing prompts that work\n- `references/media-inputs.md` — image/video reference flows\n- `references/troubleshooting.md` — common errors and fixes\n- `references/marketing-avatars.md` — preset vs custom avatars\n- `references/marketing-products.md` — URL fetch vs manual product create\n- `references/marketing-modes.md` — every Marketing Studio mode\n\nFile v1.0.0:_meta.json\n\n{\n  \"ownerId\": \"kn75nbcv2e6k6p0qdc2tkn0dvn862nys\",\n  \"slug\": \"higgsfield-generate\",\n  \"version\": \"1.0.0\",\n  \"publishedAt\": 1777911637784\n}\n\nFile v1.0.0:references/marketing-avatars.md\n\n# Avatars\n\n## Preset vs Custom\n\n| | Preset | Custom |\n|---|---|---|\n| Source | Curated by Higgsfield | User-uploaded |\n| Cost | None for selection | Cost of upload |\n| Diversity | Limited but professional | Unlimited |\n| Use when | Generic ad, fast turnaround | Brand-specific face, founder, employee |\n\n## Listing presets\n\n```bash\nhiggsfield marketing-studio avatars list\nhiggsfield marketing-studio avatars list --json | jq '.[] | select(.gender==\"female\")'\n```\n\nFilter by `name`, `gender`, etc. on the JSON output.\n\n## Creating a custom avatar\n\n```bash\nID=$(higgsfield upload create founder.png)\nURL=$(higgsfield upload create founder.png --json | jq -r .url)   # if you need cloudfront URL\nhiggsfield marketing-studio avatars create --name \"Founder\" --image $ID --image-url $URL\n```\n\n`--image-url` is the cloudfront URL from the upload. Required by the API.\n\n## Passing to video\n\n```bash\nhiggsfield generate create marketing_studio_video \\\n  --avatars '[{\"id\":\"<avatar_id>\",\"type\":\"preset\"}]' \\\n  ... \\\n  --wait\n```\n\n`type` is `preset` for curated, `custom` for user-created.\n\nFile v1.0.0:references/marketing-modes.md\n\n# Marketing Studio Modes\n\nCanonical mode values for `marketing_studio_video` `--mode`. Mirrored from the MCP server (`fnf-mcp-server/src/provider/fnf-client/generation/marketing-studio-video.ts`). All slugs below are accepted by the API.\n\n| `--mode` slug | Human-readable label | Best for |\n|---|---|---|\n| `ugc` | UGC | Default. Casual, organic-feel content from a presenter. |\n| `ugc_how_to` | Tutorial | \"Here's how to use this.\" Tutorial / explainer. |\n| `ugc_unboxing` | Unboxing | \"Just got this in the mail.\" Unboxing reveal. |\n| `product_showcase` | Hyper Motion | Clean product highlight, polished. |\n| `product_review` | Product Review | Presenter giving an opinion on the product. |\n| `tv_spot` | TV Spot | Broadcast-style commercial. Higher production. |\n| `wild_card` | Wild Card | Experimental, model picks the vibe. |\n| `ugc_virtual_try_on` | UGC Virtual Try On | Person trying on clothing/accessories — UGC vibe. |\n| `virtual_try_on` | Pro Virtual Try On | Same but more polished, model-driven. |\n\n**Default when the user doesn't specify:** `ugc`.\n\n## Picking flow\n\n- \"Looks like a real person filmed on phone\" → `ugc` family (`ugc`, `ugc_unboxing`, `ugc_virtual_try_on`, `ugc_how_to`)\n- \"Polished broadcast commercial\" → `tv_spot`\n- \"Show the product itself, less presenter\" → `product_showcase`\n- \"Presenter giving an opinion\" → `product_review`\n- \"Try clothing on someone\" → `virtual_try_on` (polished) or `ugc_virtual_try_on` (organic feel)\n- \"Surprise me / something different\" → `wild_card`\n\n## Other parameters\n\nFor `marketing_studio_video`, the API accepts:\n\n- `aspect_ratio`: `auto`, `21:9`, `16:9`, `4:3`, `1:1`, `3:4`, `9:16` (default `16:9`).\n- `duration`: integer ≥ 4 seconds. No fixed cap; check `higgsfield model get marketing_studio_video` for the current upper bound.\n- `resolution`: `480p` or `720p` (default `720p`).\n- `generate_audio`: boolean (default `false`). Generates audio for the video.\n- `avatars`: array of `{id, type}` where `type` is `preset` or `custom`. See `marketing-avatars.md`.\n- `product_ids`: array of product UUIDs (from `higgsfield marketing-studio products fetch` or `create`). See `marketing-products.md`.\n- `medias`: optional reference images with role `image`, `start_image`, or `end_image`.\n- `feature: \"click_to_ad\"`: when generating from a single landing-page URL (Click-to-Ad flow).\n- `product: { id?, url? }`: alternative single-product reference; the URL flow auto-fetches.\n\n## URL-driven Click-to-Ad shortcut\n\nFor `marketing_studio_video` driven by a product URL (no manual product create / fetch), the MCP-side flow is:\n\n1. Call `higgsfield marketing-studio products fetch --url <url> --wait` (or use `show_marketing_studio` widget action `fetch`).\n2. Call `higgsfield generate create marketing_studio_video --url <same url>` — the backend looks up / reuses the entity and submits.\n\nRepeated fetches for the same URL dedupe — the backend reuses any existing non-failed entity.\n\nFile v1.0.0:references/marketing-products.md\n\n# Products\n\nTwo ways to register a product: URL fetch (auto-imports title, description, images) or manual (provide your own).\n\n## URL fetch (default)\n\n```bash\nID=$(higgsfield marketing-studio products fetch --url https://shop.example.com/sneakers --wait --json | jq -r .id)\n```\n\n`--wait` polls until `status` is `completed` or `failed`. Default timeout 90s.\n\nIf `failed`, check `fail_reason` — usually invalid URL or scrape blocked.\n\nApp Store URLs auto-route to `webproducts` (different endpoint). Use:\n\n```bash\nhiggsfield marketing-studio webproducts fetch --url https://apps.apple.com/... --wait\n```\n\n## Manual\n\nWhen the user has product photos and details:\n\n```bash\nA=$(higgsfield upload create shoe1.png)\nB=$(higgsfield upload create shoe2.png)\nhiggsfield marketing-studio products create \\\n  --title \"AeroRun Pro\" \\\n  --description \"Lightweight running shoe\" \\\n  --image $A --image $B\n```\n\nReturns the product entity directly (no polling needed).\n\n## Manual webproduct\n\nFor App Store / web pages without URL fetch:\n\n```bash\nhiggsfield marketing-studio webproducts create \\\n  --url \"https://example.com\" \\\n  --title \"MyApp\" \\\n  --subtitle \"Productivity for teams\" \\\n  --description \"...\" \\\n  --favicon-url \"https://example.com/favicon.png\" \\\n  --desktop \"https://cdn/screenshot1.png\" \\\n  --mobile \"https://cdn/mobile-screenshot.png\"\n```\n\n## Listing\n\n```bash\nhiggsfield marketing-studio products list\nhiggsfield marketing-studio webproducts list\n```\n\nFile v1.0.0:references/media-inputs.md\n\n# Media Inputs\n\nHow to pass reference images, videos, and audio. Mirrored from MCP server media-handling logic.\n\n## Path or UUID — both work\n\nEach media flag accepts either a local file path or a UUID. The CLI auto-uploads paths before submission and auto-detects whether a UUID is an upload id (from `higgsfield upload create`) or a previous job id.\n\n```bash\n# Local path — CLI uploads automatically\nhiggsfield generate create nano_banana_2 --prompt \"stylize in watercolor\" --image ./photo.png --wait\n\n# Upload id (from higgsfield upload create)\nhiggsfield generate create nano_banana_2 --prompt \"...\" --image <upload_id> --wait\n\n# Job id from a previous generation\nhiggsfield generate create seedance_2_0 --prompt \"anim\" --start-image <previous_job_id> --wait\n```\n\nType auto-detected from extension:\n\n- Image: `png`, `jpg`/`jpeg`, `webp`, `gif`\n- Video: `mp4`, `mov`, `webm`\n- Audio: `mp3`, `wav`, `m4a`, `ogg`\n\n## Roles by model family\n\nEach model declares a closed set of accepted roles via `MEDIA_ROLES`. Pass the right role; the CLI rejects unknown ones locally before submission.\n\n| Model | Accepted roles | Notes |\n|---|---|---|\n| Most image models (`nano_banana_2`, `flux_2`, `seedream_v4_5`, `gpt_image_2`, …) | `image` | 1+ references, often up to 8. |\n| `seedance_2_0` | `image`, `start_image`, `end_image`, `video`, `audio` | Audio is via `medias` (role `audio`), NOT via `--generate-audio`. |\n| `kling3_0` | `start_image`, `end_image` | Image-to-video with optional last-frame transition. |\n| `kling2_6` | `start_image` | Single frame anchor. |\n| `veo3_1` | `start_image` | Max 1 reference. |\n| `veo3` | `image` | Single image-to-video. |\n| `marketing_studio_video` | `image`, `start_image`, `end_image` | Plus `avatars`, `product_ids`, `assets` as separate fields. |\n| `z_image`, `soul_cast`, `soul_location` | (none) | Text-only. Reject media inputs. |\n\nFor simple image-to-video on a video model that only declares `image` (e.g. `veo3`), plain `--image` is auto-remapped to `start_image` by the CLI when unambiguous. When in doubt:\n\n```bash\nhiggsfield model get <model_id>   # shows the accepted media roles for this model\n```\n\n## Multiple images\n\nMost image models accept multiple references — repeat the `--image` flag:\n\n```bash\nhiggsfield generate create nano_banana_2 --prompt \"...\" \\\n  --image ./a.png --image ./b.png --image <upload_id> \\\n  --wait\n```\n\nSingle-reference video models (`veo3`, `veo3_1`, `kling2_6`) reject extra images — the CLI errors locally before submission with `Model accepts only one image reference`.\n\n## Audio reference (Seedance)\n\n`seedance_2_0` is the one model that takes an audio reference for lipsync / soundtrack matching. Pass via `medias` with role `audio`:\n\n```bash\nhiggsfield generate create seedance_2_0 \\\n  --prompt \"person speaking\" \\\n  --start-image ./headshot.png \\\n  --audio ./voice.mp3 \\\n  --duration 8 \\\n  --wait\n```\n\n**Do NOT pass `--generate-audio` to `seedance_2_0`** — the model schema doesn't declare it. Use the audio media role instead.\n\n## Schema mismatches\n\nThe CLI returns specific error messages for known shape mismatches:\n\n- `Model accepts only --image (no roles)` — the model uses the legacy `input_images` shape, not `medias` with roles. Drop role-prefixed flags and use plain `--image`.\n- `Model does not accept media inputs` — the model is text-only (`z_image`, `soul_location`, `soul_cast`, `wan2_6` for some configs). Drop all media flags.\n- `Unknown media role \"<role>\"` — the role isn't in this model's `MEDIA_ROLES`. Run `higgsfield model get <model>` and check `medias[].roles`.\n\n## Seeing what a model accepts\n\n```bash\nhiggsfield model get <model_id> --json | jq '{aspect_ratios, durations, parameters, medias}'\n```\n\nReturns the full schema: aspect ratios (closed enum or open), durations (closed list or `min/max` range), parameters (with descriptions and defaults), and media roles per slot.\n\nFile v1.0.0:references/model-catalog.md\n\n# Model Catalog\n\nThe full lineup of generation models available through Higgsfield. Each entry has its own sweet spot — pick the one that matches your brief. For the actual `--model` ID to pass to `higgsfield generate create`, run `higgsfield model list --json` and look up by display name.\n\nPreferred defaults for examples and quick-start guidance in this repo:\n- **Images:** `gpt_image_2` (general/high-fidelity) and `nano_banana_2` (character/cartoon).\n- **Video:** `seedance_2_0` (all-purpose serious video).\n\n---\n\n## Image models\n\n| Model | Provider | What it's for |\n|---|---|---|\n| Nano Banana 2 | Google | **Fast everyday default for character work.** Edits, general generation, character / cartoon / animated-style outputs. The reach-for-this model when the brief calls for character or cartoon-style image generation. |\n| Nano Banana Pro | Google | **Top-tier Nano Banana.** Same canvas as Nano Banana 2 with extra fidelity and accuracy on harder briefs. Pick when 2 isn't getting there. |\n| Nano Banana | Google | Reliable, budget-friendly entry in the Nano Banana family — picks up the same realistic look at a lighter price point. |\n| Higgsfield Soul 2.0 | Higgsfield | **Aesthetic UGC, fashion editorial, character generation.** When the brief leans editorial, lifestyle, or \"looks like a magazine cover\". Soul-aware (accepts a Soul Character reference). |\n| Soul Cinema | Higgsfield | **Cinematic stills, film-grade lighting.** The pick when the user asks for \"cinematic\" or wants concept-art mood. |\n| Soul Cast | Higgsfield | **Distinctive, characterful personas.** When the brief calls for a creative, expressive character rather than photoreal default. Text-only (no reference image). |\n| Soul Location | Higgsfield | **Best-in-class environments and locations.** Unmatched for pure scene and place generation without a person in frame. |\n| Seedream 4.5 | Bytedance | **Vector illustrations and complex scene edits with faces.** When the brief is a face-anchored photo edit into a complex new scene (more than an outfit change), without heavy filters. |\n| Seedream 5.0 Lite | Bytedance | Same Seedream lineage as 4.5 with faster turnaround for visual-reasoning and instruction-based edits. |\n| Z Image | Tongyi-MAI | **Fastest in the catalog.** Built for speed, drafts, and LoRA-driven stylization. The pick when the brief is \"fast and cheap, let me iterate\". |\n| Flux 2.0 | Black Forest Labs | Precise prompt adherence with multiple variants (pro, flex, max). A strong creative alternative when the user wants a different look from the Banana family. |\n| Flux Kontext Max | Black Forest Labs | **Context-aware editing and style transfer.** Strong for anime, stylized looks, typography remix — when defaults feel too generic. |\n| Kling O1 Image | Kling | Versatile photorealistic image generation with broad aspect-ratio support. |\n| GPT Image 1.5 | OpenAI | Earlier-generation OpenAI image model with editing and text-rendering capabilities. |\n| GPT Image 2 | OpenAI | **Default high-fidelity image generation.** Graphic design, UI, banners, typography, and any brief with on-image text. Used by `higgsfield-product-photoshoot` under the hood. |\n| Grok Imagine | xAI | Expressive, high-contrast, bold creative outputs. Worth trying for anime and stylized looks. |\n| Cinema Studio Image 2.5 | Higgsfield | Cinematic still frames up to 4K, dramatic film look. |\n| Marketing Studio Image | Higgsfield | **Branded image ads.** Retrieval-augmented over the user's avatars and products — runs inside the Marketing Studio flow. |\n| Auto | Higgsfield | **Smart routing layer.** Picks the best image model from the prompt automatically. Use when the user's intent is open and you don't want to commit to a specific model. |\n\n## Video models\n\n| Model | Provider | What it's for |\n|---|---|---|\n| Seedance 2.0 | Bytedance | **SOTA all-purpose video.** Crisp, consistent identity, multi-shot capable. The default for any serious motion / cinematic / production brief. |\n| Kling 3.0 | Kling | **Cheaper Seedance 2.0 substitute** for single-plane scenes that don't need heavy motion. Multi-shot, audio sync, motion transfer. |\n| Seedance 1.5 Pro | Bytedance | A budget-friendly Seedance for clean single-take shots. |\n| Marketing Studio | Higgsfield | **All advertising and commercial video** — UGC, unboxing, TV spot, product showcase. The default whenever the brief is \"make an ad\". See `marketing-modes.md`. |\n| Cinema Studio Video 3.0 | Higgsfield | **Top-tier cinema-grade execution.** The pick for film-look briefs at the highest fidelity. |\n| Veo 3.1 Lite | Google | **Fast and cost-effective Veo.** Built for batch and volume work. |\n| Google Veo 3.1 | Google | Ultra-realistic, top-tier cinematic quality. Quality tiers basic/high/ultra. Format set is constrained — verify accepted aspect ratio and duration before submitting. |\n| Google Veo 3 | Google | Reliable cinematic with broad creative range and audio support. |\n| Minimax Hailuo | Hailuo | **Cheap with strong physics.** Solid budget pick when natural-physics motion matters; no audio in current variants. |\n| Wan 2.7 | Wan | Synchronized audio with character-consistent video. The newer Wan release. |\n| Wan 2.6 | Wan | Open-weight, stylized, experimental creative. Cheap option when the brief is intentionally artistic. |\n| Kling 2.6 | Kling | Cinematic motion with advanced physics — earlier Kling release alongside 3.0. |\n| Grok Imagine (video) | xAI | Text and image-to-video with audio support. Worth trying for stylized creative briefs. |\n| Cinema Studio Video | Higgsfield | Cinematic compositions with dramatic mood. Use Cinema Studio Video 3.0 as the modern default. |\n| Cinema Studio Video v2 | Higgsfield | Refined cinematic camera and color with genre control. Use Cinema Studio Video 3.0 as the modern default. |\n\n---\n\n## Picking flow\n\nPractical defaults from production use. Match by intent, not surface keyword. When two could apply, the higher entry wins.\n\n### Image — pick this default\n\n1. **Brand product visual (Pinterest pin, lifestyle, hero banner, ad pack, virtual try-on, restyle)** → use `higgsfield-product-photoshoot` instead. NOT this skill.\n2. **Branded ad image with presenter avatar + product (Marketing Studio shape with RAG over user assets)** → Marketing Studio Image.\n3. **Aesthetic UGC / fashion editorial / lifestyle character** → Soul 2.0.\n4. **Cinematic still frame** → Soul Cinema.\n5. **Highly characterful, creative character (text-only, distinctive persona, no reference photo)** → Soul Cast.\n6. **Locations / environments / no-people scenes** → Soul Location. Best in class — nothing else matches.\n7. **Vector illustrations OR face edit + complex scene swap (more than outfit change, no heavy filters)** → Seedream 4.5. Seedream 5.0 Lite for the same niche but faster.\n8. **Soul Character (reference id from `higgsfield-soul-id`)** → Soul 2.0 for stills; Soul Cinema for cinematic vibe.\n9. **Anime / stylized / non-default look where defaults feel flat** → Flux Kontext Max or Grok Imagine. Worth trying.\n10. **Character or cartoon-style work** → Nano Banana 2; step up to Nano Banana Pro on hard cases.\n11. **Fast and cheap iteration / drafts / LoRA work** → Z Image.\n12. **Default for everything else** → GPT Image 2. High-fidelity general generation, graphic design, UI, banners, anything with on-image text.\n13. **Intent-only request, no preference, want auto-routing** → Auto.\n\n### Video — pick this default\n\n1. **All advertising / commercial video (UGC, unboxing, TV spot, product showcase, branded ad)** → Marketing Studio. See `marketing-modes.md`.\n2. **Default all-purpose serious video (multi-shot, consistent identity, motion-heavy, production work)** → Seedance 2.0. SOTA.\n3. **Single-plane scene without strong dynamics, cheaper** → Kling 3.0. Substitute for Seedance 2.0 when motion isn't critical.\n4. **Cheap clean shot without cuts** → Seedance 1.5 Pro.\n5. **Image-to-video with explicit first frame** → Kling 3.0 with a start frame, or Seedance 2.0 with a start frame for higher motion.\n6. **Cinema-grade execution (highest fidelity, film look)** → Cinema Studio Video 3.0.\n7. **Cheap with strong physics, audio not needed** → Minimax Hailuo.\n8. **Fast batch / volume** → Veo 3.1 Lite.\n9. **Veo-format-bound work (specific aspect / duration set Veo accepts)** → Veo 3.1; Veo 3 is slightly behind.\n10. **Stylized / animation-style edit-driven work** → Wan 2.7.\n11. **Stylized cheap experimental** → Wan 2.6.\n12. **Anime / bold-style outputs where defaults feel flat** → Grok Imagine (video). Worth trying.\n\n### Things to keep in mind\n\n- **Don't invent model names.** Run `higgsfield model list` if you're unsure — submitting an unknown model returns `unknown model \"...\"`.\n- **Audio reference for Seedance 2.0** comes through the media inputs with role `audio`, not via a separate `generate_audio` flag.\n- **Text-only models reject reference images.** Z Image, Soul Cast, Soul Location, and some Wan configs are text-only; pass no media flags to them.\n- **Route branded product visuals through `higgsfield-product-photoshoot`** — its prompt enhancer adds 10 mode-specific templates on top of GPT Image 2. Direct GPT Image 2 generation here is the right call for everything that isn't a product photoshoot.\n- **For cinema video, prefer Cinema Studio Video 3.0** as the modern default; reach for the earlier Cinema Studio Video variants only when the user names them.\n- **When the user names a specific model, use it.** The defaults above cover the common intents — the rest of the catalog exists for users who know what they want.\n\n---\n\n## Media role conventions\n\nEach model accepts a fixed set of media roles. When unsure, run `higgsfield model get <model>` and inspect the `medias[].roles` field.\n\n| Model | Accepted media roles |\n|---|---|\n| Seedance 2.0 | `image`, `start_image`, `end_image`, `video`, `audio` |\n| Kling 3.0 | `start_image`, `end_image` |\n| Kling 2.6 | `start_image` |\n| Veo 3.1 | `start_image` (max 1) |\n| Veo 3 | `image` (max 1) |\n| Marketing Studio (video) | `image`, `start_image`, `end_image` |\n| Most image models | `image` (1+) |\n| Z Image, Soul Cast, Soul Location | (no media — text-only) |\n\nFor simple image-to-video, the `start_image` role is what you want. For pure video models that only declare `image`, the `image` flag is auto-remapped to `start_image` by the CLI.\n\n## Aspect ratios and durations\n\nThese are model-specific. The CLI clamps unsupported values to the nearest allowed one (with a `Note: adjustments applied` warning) when the model declares a closed set. When in doubt:\n\n```bash\nhiggsfield model get <model>\n```\n\nCommon patterns:\n\n- **Seedance 2.0** image: `auto`, `21:9`, `16:9`, `4:3`, `1:1`, `3:4`, `9:16`. Duration 4–15s.\n- **Kling 3.0**: `16:9`, `9:16`, `1:1`. Duration 3–15s. Modes `pro`/`std`.\n- **Veo 3.1**: `16:9` or `9:16`. Duration `4`, `6`, or `8` only. Quality `basic`/`high`/`ultra`.\n- **Marketing Studio (video)**: `auto`/`21:9`/`16:9`/`4:3`/`1:1`/`3:4`/`9:16`. Resolution `480p` or `720p`.\n\n## When you submit an unknown value\n\nThe CLI reports two kinds of feedback:\n\n- **Adjustments** — a non-fatal coercion. E.g. you passed `aspect_ratio=99:99` and the model accepts a closed set; the CLI picks the closest match and continues. The adjustments map is included in the response.\n- **Validation error** — a fatal mismatch. E.g. an unknown declared parameter, or a media role the model doesn't accept. The CLI returns an error and does not submit.\n\nFile v1.0.0:references/prompt-engineering.md\n\n# Prompt Engineering\n\n## Basics\n\nHiggsfield models reward concrete, sensory prompts.\n\n- **Subject + setting + style**: \"a red fox curled in a snowy pine forest, golden hour, cinematic\"\n- **Camera**: lens (35mm, 85mm), angle (low, overhead), motion (dolly in, tracking shot)\n- **Lighting**: rim light, neon glow, moody backlight\n- **Style/medium**: oil painting, watercolor, photograph, anime, 3D render\n\nKeep it under ~200 tokens. Models distort with very long prompts.\n\n## Image-to-image\n\nWhen passing `--image`, the prompt should describe what changes, not redescribe the input.\n\nBad: \"a man with brown hair in a leather jacket holding coffee, made into anime\"\nGood: \"transform into anime style, vibrant colors, soft cel shading\"\n\n## Image-to-video\n\n`--start-image` anchors the first frame. Prompt describes motion.\n\n- Verbs: zooms in, dollies left, sweeping pan, slow push, fast whip\n- Subject motion: \"the dancer spins\", \"smoke rises slowly\"\n- Don't redescribe the static frame — model already has it.\n\n## Negative phrasing\n\nMost models don't expose a `negative_prompt`. Phrase positively:\n- Instead of \"no blur\" → \"tack sharp\"\n- Instead of \"no people\" → \"uninhabited landscape\"\n\n## Aspect ratio guidance\n\n- `16:9` — landscape, cinematic\n- `9:16` — vertical, social\n- `1:1` — square, profile / icon\n- `4:3`, `3:4`, `21:9` — model-dependent, check `higgsfield model get <jst>`\n\n## Safety\n\nModels reject prompts with `nsfw` or `ip_detected` terminal status. Avoid:\n- Real public figures\n- Sexual content\n- Trademarks / branded characters\n\nFile v1.0.0:references/troubleshooting.md\n\n# Troubleshooting\n\n## Authentication\n\n- `Session expired.` → `higgsfield auth login`\n- `Stored credentials are for ... but current environment ...` → `higgsfield auth login` for the current API URL.\n- `Not authenticated.` → first `higgsfield auth login`.\n\n## Validation\n\n- `Missing required params: prompt` — user gave no prompt. Ask.\n- `Invalid values: <param>=<v> (allowed: ...)` — pick from allowed enum.\n- `Unknown params: <name>` — schema doesn't accept this flag. Run `higgsfield model get <jst>` and check.\n\n## Job lifecycle\n\n- `Job ended with status \"failed\"` — server-side failure. Often prompt content / safety. Try rephrasing.\n- `nsfw` / `ip_detected` — content policy. Rephrase.\n- `Timeout after 10m` — model is slow today. Bump `--timeout 30m` or retry.\n\n## Rate limits\n\n`Higgsfield API error (HTTP 429)` — too many requests. Back off.\n\n## CloudFlare / DataDome\n\nIf `Failed to decode response. Body: <html>...captcha-delivery...` appears, the server's anti-bot fired. Wait 30s and retry. If persistent, ping the team.\n\n## Cost\n\n`higgsfield generate cost <jst> ...` returns credit estimate without submitting. Useful when the user asks \"how much will this cost?\".\n\nFile v1.0.0:skill-card.md\n\n## Description:\n\nGenerate images and videos with Higgsfield AI models, including generic image/video generation and Marketing Studio workflows for branded ad images and videos with avatars and products.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[higgsfield](https://clawhub.ai/user/higgsfield)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nExternal users, creators, marketers, and developers use this skill to ask an agent to select Higgsfield models, prepare CLI commands, submit image or video generation jobs, and return the generated media URL with a brief summary.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: The skill may send prompts, media files, product URLs, and related assets to Higgsfield.\n\nMitigation: Use only with content the user is comfortable sharing with Higgsfield, and avoid sensitive or restricted assets unless the user has explicitly approved that use.\n\nRisk: The skill instructs the agent to install the Higgsfield CLI with an unverified remote installer script.\n\nMitigation: Do not run the remote installer automatically; install the CLI through a pinned, trusted, inspectable process before using the skill.\n\nRisk: The skill requires Higgsfield account authentication before jobs can be submitted.\n\nMitigation: Have the user authenticate intentionally with `higgsfield auth login` and confirm completion before the agent continues.\n\n## Reference(s):\n\n- [Model Catalog](references/model-catalog.md)\n- [Media Inputs](references/media-inputs.md)\n- [Marketing Studio Modes](references/marketing-modes.md)\n- [Marketing Studio Avatars](references/marketing-avatars.md)\n- [Marketing Studio Products](references/marketing-products.md)\n- [Prompt Engineering](references/prompt-engineering.md)\n- [Troubleshooting](references/troubleshooting.md)\n- [ClawHub skill page](https://clawhub.ai/higgsfield/skills/higgsfield-generate)\n\n## Skill Output:\n\n**Output Type(s):** [text, shell commands, guidance]\n\n**Output Format:** [Markdown text with inline shell commands and generated media URLs]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [May invoke the Higgsfield CLI to submit generation jobs and return result URLs after waiting for completion.]\n\n## Skill Version(s):\n\n1.0.0 (source: server release metadata; artifact frontmatter version: 0.3.0)\n\n## Ethical Considerations:\n\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.","readmeExcerpt":"Skill: Higgsfield Generate Owner: higgsfield Summary: Generate images and videos via Higgsfield AI through 30+ models including Nano Banana 2, Soul V2, Veo 3.1, Kling 3.0, Seedance 2.0, Flux 2, GPT Image 2, plus... Tags: latest:1.0.0 Version history: v1.0.0 | 2026-05-04T16:20:37.784Z | user - Initial release of higgsfield-generate for image and video generation via Higgsfield AI, supporting 30+ models. - Supports a w","codeSnippets":[],"executableExamples":[{"language":"bash","snippet":"curl -fsSL https://raw.githubusercontent.com/higgsfield-ai/cli/main/install.sh | sh"},{"language":"bash","snippet":"curl -fsSL https://raw.githubusercontent.com/higgsfield-ai/cli/main/install.sh | sh"},{"language":"bash","snippet":"higgsfield generate create gpt_image_2 --prompt \"neon city at dusk\" --aspect_ratio 16:9 --resolution 2k --wait\nhiggsfield generate create nano_banana_2 --prompt \"anime character concept, expressive pose\" --image ./ref.png --wait\nhiggsfield generate create seedance_2_0 --prompt \"camera dollies in\" --start-image ./first.png --duration 8 --wait\nhiggsfield generate create text2image_soul_v2 --prompt \"...\" --soul-id <soul_ref_id> --wait"},{"language":"bash","snippet":"higgsfield generate create marketing_studio_video \\\n     --prompt \"...\" \\\n     --avatars '[{\"id\":\"<avatar_id>\",\"type\":\"preset\"}]' \\\n     --product_ids '[<product_id>]' \\\n     --mode ugc \\\n     --duration 15 \\\n     --resolution 720p \\\n     --aspect_ratio 9:16 \\\n     --wait"},{"language":"bash","snippet":"# 1. Trigger fetch (returns the product id and starts background scrape)\nhiggsfield marketing-studio products fetch --url https://shop.example.com/sneakers --wait\n\n# 2. Generate the marketing video against the same URL — backend reuses the entity\nhiggsfield generate create marketing_studio_video \\\n  --url https://shop.example.com/sneakers \\\n  --mode ugc \\\n  --duration 15 \\\n  --aspect_ratio 9:16 \\\n  --wait"},{"language":"bash","snippet":"higgsfield generate create marketing_studio_image \\\n  --prompt \"...\" \\\n  --aspect_ratio 1:1 \\\n  --resolution 2k \\\n  --wait"}],"parameters":null,"dependencies":[],"permissions":[],"extractedFiles":[{"path":"SKILL.md","content":"---\nversion: 0.3.0\nname: higgsfield-generate\ndescription: |\n  Generate images and videos via Higgsfield AI through 30+ models including\n  Nano Banana 2, Soul V2, Veo 3.1, Kling 3.0, Seedance 2.0, Flux 2, GPT Image 2,\n  plus Marketing Studio for branded ad video/image with curated avatars and\n  imported products.\n  Use when: \"generate an image\", \"make a picture\", \"create artwork\",\n  \"make a video\", \"animate this photo\", \"image-to-video\", \"img2vid\",\n  \"edit this image with AI\", \"stylize a photo\", \"remix this image\",\n  \"produce a clip\", \"render a scene\", \"create an ad\", \"make a UGC video\",\n  \"generate marketing video\", \"make a product demo\", \"create unboxing\",\n  \"TV spot\", \"virtual try-on\", \"product showcase\", \"brand video\",\n  \"presenter video for product\", \"import product from URL\",\n  \"create avatar for ad\".\n  Supports text-to-image, image-to-image, image-to-video, reference-based\n  generation, and Marketing Studio (avatars + products + ad modes).\n  Auto-detects whether passed IDs are uploads or previous jobs.\n  Chain with higgsfield-soul-id when the user wants their face in the output.\n  NOT for: training Soul Character (use higgsfield-soul-id), professional product\n  photoshoots with mode-specific prompt enhancement (use\n  higgsfield-product-photoshoot), text-only / chat / TTS tasks.\nargument-hint: \"[prompt] [--model <name>] [--image <path-or-id>]\"\nallowed-tools: Bash\n---\n\n# Higgsfield Generate\n\nSubmit jobs to any Higgsfield model. Wraps the `higgsfield` CLI. Covers generic image/video gen and Marketing Studio (branded ads, avatars, products).\n\n## Step 0 — Bootstrap\n\nBefore any other command, make sure the CLI is installed and authenticated:\n\n1. If `higgsfield` is not on `$PATH`, install it:\n   ```bash\n   curl -fsSL https://raw.githubusercontent.com/higgsfield-ai/cli/main/install.sh | sh\n   ```\n2. If `higgsfield account status` fails with `Session expired` / `Not authenticated`, ask the user to run `higgsfield auth login` (interactive, opens a browser) and wait for them to confirm before continuing.\n\nSkip both checks if `higgsfield account status` already prints account info.\n\n## UX Rules\n\n1. Be concise. No raw IDs, no JSON dumps in chat. Print result URL when ready.\n2. No internal jargon. Don't narrate \"calling higgsfield cost\", \"polling job\".\n3. Detect the user's language from the first message and reply in it. Technical args (`--aspect_ratio 16:9`) stay English.\n4. Don't batch-ask. Pick a sane default model and ask one thing at a time only if genuinely missing.\n5. Don't pre-estimate cost. Just submit unless the user asks.\n6. Pass `--wait` to `generate create` so the command blocks until done and prints the result URL itself. Avoid the two-step `create` → `wait` pattern.\n\n## Workflow — generic generation\n\n1. **Pick a model.** Practical defaults from production use:\n\n   **Image:**\n   - Brand product visual (Pinterest pin, lifestyle, hero banner, ad pack, virtual try-on) → use `higgsfield-product-photoshoot` instead. NOT this skill.\n   - Branded "},{"path":"_meta.json","content":"{\n  \"ownerId\": \"kn75nbcv2e6k6p0qdc2tkn0dvn862nys\",\n  \"slug\": \"higgsfield-generate\",\n  \"version\": \"1.0.0\",\n  \"publishedAt\": 1777911637784\n}"},{"path":"references/marketing-avatars.md","content":"# Avatars\n\n## Preset vs Custom\n\n| | Preset | Custom |\n|---|---|---|\n| Source | Curated by Higgsfield | User-uploaded |\n| Cost | None for selection | Cost of upload |\n| Diversity | Limited but professional | Unlimited |\n| Use when | Generic ad, fast turnaround | Brand-specific face, founder, employee |\n\n## Listing presets\n\n```bash\nhiggsfield marketing-studio avatars list\nhiggsfield marketing-studio avatars list --json | jq '.[] | select(.gender==\"female\")'\n```\n\nFilter by `name`, `gender`, etc. on the JSON output.\n\n## Creating a custom avatar\n\n```bash\nID=$(higgsfield upload create founder.png)\nURL=$(higgsfield upload create founder.png --json | jq -r .url)   # if you need cloudfront URL\nhiggsfield marketing-studio avatars create --name \"Founder\" --image $ID --image-url $URL\n```\n\n`--image-url` is the cloudfront URL from the upload. Required by the API.\n\n## Passing to video\n\n```bash\nhiggsfield generate create marketing_studio_video \\\n  --avatars '[{\"id\":\"<avatar_id>\",\"type\":\"preset\"}]' \\\n  ... \\\n  --wait\n```\n\n`type` is `preset` for curated, `custom` for user-created."},{"path":"references/marketing-modes.md","content":"# Marketing Studio Modes\n\nCanonical mode values for `marketing_studio_video` `--mode`. Mirrored from the MCP server (`fnf-mcp-server/src/provider/fnf-client/generation/marketing-studio-video.ts`). All slugs below are accepted by the API.\n\n| `--mode` slug | Human-readable label | Best for |\n|---|---|---|\n| `ugc` | UGC | Default. Casual, organic-feel content from a presenter. |\n| `ugc_how_to` | Tutorial | \"Here's how to use this.\" Tutorial / explainer. |\n| `ugc_unboxing` | Unboxing | \"Just got this in the mail.\" Unboxing reveal. |\n| `product_showcase` | Hyper Motion | Clean product highlight, polished. |\n| `product_review` | Product Review | Presenter giving an opinion on the product. |\n| `tv_spot` | TV Spot | Broadcast-style commercial. Higher production. |\n| `wild_card` | Wild Card | Experimental, model picks the vibe. |\n| `ugc_virtual_try_on` | UGC Virtual Try On | Person trying on clothing/accessories — UGC vibe. |\n| `virtual_try_on` | Pro Virtual Try On | Same but more polished, model-driven. |\n\n**Default when the user doesn't specify:** `ugc`.\n\n## Picking flow\n\n- \"Looks like a real person filmed on phone\" → `ugc` family (`ugc`, `ugc_unboxing`, `ugc_virtual_try_on`, `ugc_how_to`)\n- \"Polished broadcast commercial\" → `tv_spot`\n- \"Show the product itself, less presenter\" → `product_showcase`\n- \"Presenter giving an opinion\" → `product_review`\n- \"Try clothing on someone\" → `virtual_try_on` (polished) or `ugc_virtual_try_on` (organic feel)\n- \"Surprise me / something different\" → `wild_card`\n\n## Other parameters\n\nFor `marketing_studio_video`, the API accepts:\n\n- `aspect_ratio`: `auto`, `21:9`, `16:9`, `4:3`, `1:1`, `3:4`, `9:16` (default `16:9`).\n- `duration`: integer ≥ 4 seconds. No fixed cap; check `higgsfield model get marketing_studio_video` for the current upper bound.\n- `resolution`: `480p` or `720p` (default `720p`).\n- `generate_audio`: boolean (default `false`). Generates audio for the video.\n- `avatars`: array of `{id, type}` where `type` is `preset` or `custom`. See `marketing-avatars.md`.\n- `product_ids`: array of product UUIDs (from `higgsfield marketing-studio products fetch` or `create`). See `marketing-products.md`.\n- `medias`: optional reference images with role `image`, `start_image`, or `end_image`.\n- `feature: \"click_to_ad\"`: when generating from a single landing-page URL (Click-to-Ad flow).\n- `product: { id?, url? }`: alternative single-product reference; the URL flow auto-fetches.\n\n## URL-driven Click-to-Ad shortcut\n\nFor `marketing_studio_video` driven by a product URL (no manual product create / fetch), the MCP-side flow is:\n\n1. Call `higgsfield marketing-studio products fetch --url <url> --wait` (or use `show_marketing_studio` widget action `fetch`).\n2. Call `higgsfield generate create marketing_studio_video --url <same url>` — the backend looks up / reuses the entity and submits.\n\nRepeated fetches for the same URL dedupe — the backend reuses any existing non-failed entity."},{"path":"references/marketing-products.md","content":"# Products\n\nTwo ways to register a product: URL fetch (auto-imports title, description, images) or manual (provide your own).\n\n## URL fetch (default)\n\n```bash\nID=$(higgsfield marketing-studio products fetch --url https://shop.example.com/sneakers --wait --json | jq -r .id)\n```\n\n`--wait` polls until `status` is `completed` or `failed`. Default timeout 90s.\n\nIf `failed`, check `fail_reason` — usually invalid URL or scrape blocked.\n\nApp Store URLs auto-route to `webproducts` (different endpoint). Use:\n\n```bash\nhiggsfield marketing-studio webproducts fetch --url https://apps.apple.com/... --wait\n```\n\n## Manual\n\nWhen the user has product photos and details:\n\n```bash\nA=$(higgsfield upload create shoe1.png)\nB=$(higgsfield upload create shoe2.png)\nhiggsfield marketing-studio products create \\\n  --title \"AeroRun Pro\" \\\n  --description \"Lightweight running shoe\" \\\n  --image $A --image $B\n```\n\nReturns the product entity directly (no polling needed).\n\n## Manual webproduct\n\nFor App Store / web pages without URL fetch:\n\n```bash\nhiggsfield marketing-studio webproducts create \\\n  --url \"https://example.com\" \\\n  --title \"MyApp\" \\\n  --subtitle \"Productivity for teams\" \\\n  --description \"...\" \\\n  --favicon-url \"https://example.com/favicon.png\" \\\n  --desktop \"https://cdn/screenshot1.png\" \\\n  --mobile \"https://cdn/mobile-screenshot.png\"\n```\n\n## Listing\n\n```bash\nhiggsfield marketing-studio products list\nhiggsfield marketing-studio webproducts list\n```"}],"languages":[],"docsSourceLabel":"CLAWHUB","editorialOverview":"Generate images and videos via Higgsfield AI through 30+ models including Nano Banana 2, Soul V2, Veo 3.1, Kling 3.0, Seedance 2.0, Flux 2, GPT Image 2, plus... Skill: Higgsfield Generate Owner: higgsfield Summary: Generate images and videos via Higgsfield AI through 30+ models including Nano Banana 2, Soul V2, Veo 3.1, Kling 3.0, Seedance 2.0, Flux 2, GPT Image 2, plus... Tags: latest:1.0.0 Version history: v1.0.0 | 2026-05-04T16:20:37.784Z | user - Initial release of higgsfield-generate for image and video generation via Higgsfield AI, supporting 30+ models. - Supports a w","editorialQuality":{"score":100,"threshold":65,"status":"ready","wordCount":1441,"uniquenessScore":49,"reasons":[]}},"media":{"evidence":{"source":"no-media","verified":false,"confidence":"low","updatedAt":"2026-10-09T17:07:33.531Z","emptyReason":"No screenshots, media assets, or demo links are available."},"primaryImageUrl":null,"mediaAssetCount":0,"assets":[],"demoUrl":null},"ownerResources":{"evidence":{"source":"unclaimed","verified":false,"confidence":"low","updatedAt":"2026-10-09T17:07:33.531Z","emptyReason":"This page has not been claimed by the agent owner."},"hasCustomPage":false,"customPageUpdatedAt":null,"customLinks":[],"structuredLinks":{"docsUrl":null,"demoUrl":null,"supportUrl":null,"pricingUrl":null,"statusUrl":null},"customPage":null},"relatedAgents":{"evidence":{"source":"protocol-neighbors","verified":false,"confidence":"medium","updatedAt":"2026-10-09T22:37:11.243Z","emptyReason":null},"items":[{"id":"8ebccd8e-3863-4187-8355-c3f14e1f9edf","entityType":"agent","canonicalPath":"/agent/iofficeai-aionui","slug":"iofficeai-aionui","name":"AionUi","description":"Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!","url":"https://github.com/iOfficeAI/AionUi","homepage":"https://www.aionui.com","source":"GITHUB_REPOS","protocols":["MCP","OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-10-09T19:11:12.944Z","createdAt":"2026-02-25T03:38:16.584Z","downloads":null},{"id":"b917f68a-ebff-438e-84f8-3f4b2494c0bc","entityType":"agent","canonicalPath":"/agent/activepieces-activepieces","slug":"activepieces-activepieces","name":"activepieces","description":"AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents","url":"https://github.com/activepieces/activepieces","homepage":"https://www.activepieces.com","source":"GITHUB_REPOS","protocols":["OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-04-15T02:22:12.426Z","createdAt":"2026-02-25T03:38:12.412Z","downloads":null},{"id":"5cb26759-3a39-483f-94cf-276a98c13bb8","entityType":"agent","canonicalPath":"/agent/cherryhq-cherry-studio","slug":"cherryhq-cherry-studio","name":"cherry-studio","description":"AI productivity studio with smart chat, autonomous agents, and 300+ assistants. Unified access to frontier LLMs","url":"https://github.com/CherryHQ/cherry-studio","homepage":"https://cherry-ai.com","source":"GITHUB_REPOS","protocols":["MCP","OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-04-11T14:38:40.986Z","createdAt":"2026-02-25T03:38:19.379Z","downloads":null},{"id":"6f6582d0-5d76-4f0f-b81d-86520247950b","entityType":"agent","canonicalPath":"/agent/copilotkit-copilotkit","slug":"copilotkit-copilotkit","name":"CopilotKit","description":"The Frontend for Agents & Generative UI. React + Angular","url":"https://github.com/CopilotKit/CopilotKit","homepage":"https://docs.copilotkit.ai","source":"GITHUB_REPOS","protocols":["OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-03-25T09:50:57.846Z","createdAt":"2026-02-25T03:39:14.617Z","downloads":null}],"links":{"hub":"/agent","source":"/agent/source/clawhub","protocols":[{"label":"OpenClaw","href":"/agent/protocol/openclew"}]}}}