nebula-nodes

FAL.ai model catalog access — image, video, audio, 3D, LoRA training, and more via fal-ai/ and bytedance/ and wan/ endpoints. Activate when the user configures any nebula node that routes through FAL (kling, sora, veo, flux, seedance, wan, luma, pixverse, ltx, moonvalley, hunyuan3d, recraft, nano-banana via FAL, and 30+ others), when the user mentions a fal-ai/ slug, or when they ask which FAL model to use for a task. Entry point for 160 curated per-model skills + category overviews at `.agents/skills/fal/`.

FAL (fal.ai) is Nebula's all-purpose generation gateway — one API key fronts 52 ready-to-use nodes for image, video, 3D, and audio generation (FLUX, GPT Image 1.5/2, Seedream, Ideogram 4 (gen + edit/remix/reframe/replace-background/character/upscale), Recraft, Sora 2, Kling, Wan, Luma Ray 2, PixVerse, LTX, Seedance, Hunyuan3D, Meshy, Stable Audio 2.5, ACE-Step, MMAudio V2, Demucs, Clarity Upscaler, SeedVR2 Video Upscale) plus the catch-all `fal-universal` slug node. Activate when the user configures any FAL node — flux-1-1-ultra, flux-2-pro, flux-schnell, fast-sdxl, flux-kontext, flux-fill-inpaint, gpt-image-1-5, gpt-image-1-5-edit, gpt-image-2-fal-generate, gpt-image-2-fal-edit, seedream-4-5, ideogram-v4, ideogram-edit, ideogram-remix, ideogram-reframe, ideogram-replace-background, ideogram-character, ideogram-upscale, recraft-v4-raster, recraft-v4-svg, sora-2, kling-v2-1, kling-v3, kling-o3, wan-2-6-t2v, wan-2-6-i2v, wan-2-6-r2v, luma-ray2-t2v, luma-ray2-i2v, luma-ray2-flash-modify, pixverse-v4-5, ltx-video

Anthropic (Claude) Messages API in Nebula — a single text-gen node (Claude / claude-chat) that turns a text prompt plus optional image(s) into streamed text, with system prompt, temperature/top_p/stop sampling, and extended-thinking. Use for captioning, summarizing, translating, rewriting, reasoning, and drafting prompts to feed other media nodes. Activate when the user configures the `claude-chat` node, wires a Claude/Anthropic node into a graph, or asks about Anthropic/Claude in Nebula. Sourced from the official Anthropic Messages API (platform.claude.com/docs/en/api/messages) and the Nebula audit guide docs/api-guides/anthropic.md on 2026-06-04.

ElevenLabs audio API in Nebula — text-to-speech, sound-effects-from-text, speech-to-speech voice changing, audio isolation (denoise), translated dubbing, and speech-to-text transcription, all as audio-gen nodes. Activate when the user configures any elevenlabs-tts, elevenlabs-sfx, elevenlabs-sts, elevenlabs-isolation, elevenlabs-dubbing, or elevenlabs-stt node, or asks about ElevenLabs in Nebula. Sourced from the official ElevenLabs API reference (elevenlabs.io/docs/api-reference) cross-checked against backend/handlers/elevenlabs.py + backend/data/node_definitions.json, and the Nebula audit guide docs/api-guides/elevenlabs.md, on 2026-06-05.

Best practices for prompting Google's Gemini-family APIs — Gemini text models, Imagen, Nano Banana (Gemini 3.1 Flash Image), and Veo video. Activate when writing prompts, building graphs, or configuring nodes that target any Gemini model, when editing files importing `google-genai` / `@google/genai`, or when the user asks about Imagen, Nano Banana, or Veo. Routes to topic files for deep guidance. Sourced from ai.google.dev with fetch dates noted — verify against the live docs for anything that smells stale.

Higgsfield video generation in Nebula — text-to-video and image-to-video short cinematic clips via the DoP ("Depth of Presence") model plus Kling v2.1 Pro and Seedance v1 Pro guest models, with duration and aspect-ratio control. Activate when the user configures the `higgsfield` node or asks about Higgsfield in Nebula. The node is video-output only (no Soul images, Speak lipsync, or motion-camera presets are wired). Sourced 2026-06-04 from the official Higgsfield API docs (docs.higgsfield.ai) plus the Nebula audit guide docs/api-guides/higgsfield.md.

Ideogram image nodes in Nebula — v4 generate, masked inpaint (ideogram-edit), maskless edit-prompt, remix, reframe, replace-background, character, upscale, transparent, remove-background, describe, magic-prompt. Activate when the user configures any ideogram-* node, asks about Ideogram in Nebula, or wants to wire or test mask-painter → ideogram-edit inpainting on the canvas. Covers dual-route auth (IDEOGRAM_API_KEY direct vs FAL_KEY fallback), mask polarity (BLACK = edit), the four-wire canvas graph, Inspector mask painting, expand_prompt, preset image execution, auto-wire of base image, and how to verify a visible inpaint result. Sourced from docs/api-guides/ideogram.md and live canvas E2E on feat/inpainting (2026-06-10).

MiniMax (Hailuo) video generation in Nebula — text-to-video, image-to-video (first frame), and subject-reference (face-consistent) video via Hailuo 2.3 / Hailuo 2.3 Fast / Hailuo 02 / S2V-01 on the async /v1/video_generation endpoint. VIDEO ONLY in Nebula (the API also offers TTS, voice clone, music, and image gen, but no Nebula node wires those). Activate when the user configures any minimax-t2v, minimax-i2v, or minimax-s2v node, or asks about MiniMax / Hailuo in Nebula. Sourced from the official MiniMax platform docs (platform.minimax.io) and the Nebula audit guide docs/api-guides/minimax.md, cross-checked against backend/handlers/minimax.py and node_definitions.json on 2026-06-04 (model enums updated 2026-06-10).

Create polished motion-design videos for Nebula Nodes — product announcements, feature showcases, demo clips, launch/promo videos — by capturing the REAL Slava UI live and compositing it with HyperFrames (GSAP) using a Firecrawl-style motion grammar and a render → inspect → fix loop. Use this whenever the user wants a Nebula Nodes video, announcement, showcase, demo, promo, or "show off / make a clip for [feature]"; whenever editing a HyperFrames composition under hyperframes/* for Nebula; or when capturing the live Nebula app UI for a video. Covers the capture pipeline (driving the window.__nebula* stores), HyperFrames + GSAP gotchas (hard cuts vs cross-fades, deterministic timeline, typing/caret, centered-card transforms), the Slava brand + motion grammar, and the readability-first quality rubric. Reach for this even if the user doesn't say "HyperFrames."

Nous Portal (Hermes) provider in Nebula — a single universal text/reasoning LLM node backed by Nous Research's open-weight Hermes models (Hermes-4.3-36B, Hermes-4-70B, Hermes-4-405B, all 128k context), OpenAI-compatible SSE streaming, long-context drafting/summarization/chain-of-thought. Unusual auth — NO env var; the handler reads an OAuth Bearer from ~/.hermes/auth.json. Text-only — no tools, no JSON mode, no Tool Gateway. Activate when the user configures the `nous-portal-universal` node or asks about Nous Portal / Hermes in Nebula. Sourced 2026-06-04 from the Nous Portal API docs (https://portal.nousresearch.com/api-docs) + the Nebula audit guide docs/api-guides/nous.md.

OpenAI (BYOK, direct) in Nebula — text-to-image (GPT Image 1/1.5/mini, GPT Image 2 up to 4K), image edit + inpaint (single image or up to 10 refs, optional mask), text-to-speech (13 voices, 6 formats, voice instructions), Whisper/GPT-4o transcription (text/json/srt/vtt), audio translation to English, and OpenAI Chat (GPT-5.5/5.4 + legacy GPT-4o/4.1) with vision and JSON-object output. Activate when the user configures any gpt-image-1-generate, gpt-image-1-edit, gpt-image-2-generate, gpt-image-2-edit, openai-tts, openai-stt, openai-translate, or gpt-4o-chat node, or asks about OpenAI in Nebula. For gpt-image-2 prompting craft and the FAL-routed gpt-image-2 nodes, see .agents/skills/gpt-image-2/SKILL.md. Sourced from the official OpenAI API docs (platform.openai.com / developers.openai.com) and the Nebula audit guide docs/api-guides/openai.md, verified against the backend handlers on 2026-06-04 (updated 2026-06-10).

OpenRouter universal gateway in Nebula — one node (openrouter-universal) that reaches hundreds of models (GPT, Claude, Gemini, Llama, DeepSeek, FLUX, Recraft, and more) through OpenRouter's chat-completions API. Capabilities exposed in Nebula: streaming text/chat/reasoning/coding, vision (image-in → text-out), and image generation (text → image) — the model you pick decides which. Activate when the user configures the `openrouter-universal` node (display name "OpenRouter"), picks/searches an OpenRouter model in the Inspector, or asks about OpenRouter in Nebula. Sourced from the official OpenRouter docs (openrouter.ai/docs) and the Nebula audit guide docs/api-guides/openrouter.md on 2026-06-04 — every node id, port, param, header, and capability boundary here is cross-checked against backend/handlers/openrouter.py, backend/routes/openrouter_proxy.py, and backend/data/node_definitions.json.

QuiverAI (Arrow) SVG API — text-to-SVG generation (icons, logos, flat illustrations) and raster-to-SVG vectorization, with reference-image guidance (up to 16), multi-output, sampling controls, and live SSE streaming of draft SVGs. Activate when the user configures a `quiver-arrow-generate` or `quiver-arrow-vectorize` node, or asks about QuiverAI / Arrow / "vectorize" / "text to SVG" / "generate an SVG logo" in Nebula. Sourced 2026-06-04 from the official QuiverAI API docs (docs.quiver.ai) and the Nebula audit guide docs/api-guides/quiver.md, cross-checked against backend/data/node_definitions.json, backend/handlers/quiver.py, and backend/services/quiver_client.py.

Replicate universal gateway — run any Replicate-hosted model (image, video, audio, 3D, or text) by typing its owner/name slug into one node, with one API token. Capabilities: text-to-image, image editing/upscaling, text/image-to-video, music/TTS/transcription, image/text-to-3D, and open LLM text — all via a single passthrough node that resolves the model version, creates an async prediction, and polls to completion. Activate when the user configures the `replicate-universal` node (display name "Replicate") or asks about Replicate in Nebula. Sourced from the official Replicate HTTP API reference (replicate.com/docs/reference/http) and the Nebula audit guide docs/api-guides/replicate.md on 2026-06-04 — node id, the single model_id param, async-poll execution, and output-type inference are cross-checked against backend/data/node_definitions.json and backend/handlers/replicate_universal.py.

xAI (Grok Imagine) video generation in Nebula — text-to-video and image-to-video short clips with duration, aspect-ratio, and resolution control via the Grok Imagine Video model. Activate when the user configures the `grok-imagine-video` node or asks about xAI / Grok / Grok Imagine in Nebula. Sourced from the official xAI docs (docs.x.ai — Imagine video generation reference) and the Nebula audit guide docs/api-guides/xai.md on 2026-06-04.

OpenAI (BYOK, direct) in Nebula — text-to-image (GPT Image 1/1.5/mini, GPT Image 2 up to 4K), image edit + inpaint (single image or up to 10 refs, optional mask), text-to-speech (13 voices, 6 formats, voice instructions), Whisper/GPT-4o transcription (text/json/srt/vtt), audio translation to English, and OpenAI Chat (GPT-5.5/5.4 + legacy GPT-4o/4.1) with vision and JSON-object output. Activate when the user configures any gpt-image-1-generate, gpt-image-1-edit, gpt-image-2-generate, gpt-image-2-edit, openai-tts, openai-stt, openai-translate, or gpt-4o-chat node, or asks about OpenAI in Nebula. For gpt-image-2 prompting craft and the FAL-routed gpt-image-2 nodes, see .claude/skills/gpt-image-2/SKILL.md. Sourced from the official OpenAI API docs (platform.openai.com / developers.openai.com) and the Nebula audit guide docs/api-guides/openai.md, verified against the backend handlers on 2026-06-04 (updated 2026-06-10).

Ideogram image nodes in Nebula — v4 generate, masked inpaint (ideogram-edit), maskless edit-prompt, remix, reframe, replace-background, character, upscale, transparent, remove-background, describe, magic-prompt. Activate when the user configures any ideogram-* node, asks about Ideogram in Nebula, or wants to wire or test mask-painter → ideogram-edit inpainting on the canvas. Covers dual-route auth (IDEOGRAM_API_KEY direct vs FAL_KEY fallback), mask polarity (BLACK = edit), the four-wire canvas graph, Inspector mask painting, expand_prompt, preset image execution, auto-wire of base image, and how to verify a visible inpaint result. Sourced from docs/api-guides/ideogram.md and live canvas E2E on feat/inpainting (2026-06-10).

Google's Gemini / Imagen / Veo / Lyria models inside Nebula — text & vision chat, Imagen 4 text-to-image, Nano Banana image gen/edit/compose, Veo 3.1 video with sound, Lyria 3 music, Gemini TTS narration, text embeddings, and a Style Reference utility. Activate when the user configures any gemini-chat, imagen-4-generate, nano-banana, veo-3, lyria-3, gemini-tts, gemini-embeddings, or style-reference node, or asks about Gemini / Imagen / Veo / Nano Banana / Lyria in Nebula. For prompt craft also load the topic files (nano-banana.md, veo.md, imagen.md, gemini-text.md). Sourced from ai.google.dev (image-generation, imagen, video, speech-generation, music-generation, embeddings) and the Nebula audit guide docs/api-guides/google.md, cross-checked against backend/data/node_definitions.json + backend/handlers/google_gemini.py + backend/handlers/veo.py on 2026-06-04.

Runway API in Nebula — image-to-video & text-to-video (Gen-4.5, Seedance 2.0 (+ Fast), HappyHorse 1.0, Gen-4 Turbo, Gen-3a Turbo, Veo 3.1, Veo 3.1 Fast, Veo 3), Aleph 2.0 / Gen-4 Aleph video restyle, Gen-4 Image / Nano Banana Pro / GPT Image 2 / Gemini 2.5 Flash stills, Magnific-backed image upscaling, Act-Two character performance, and ElevenLabs-backed TTS / speech-to-speech / voice dubbing. Activate when the user configures any of the Nebula nodes runway-video, runway-aleph, runway-image, runway-upscale, runway-act-two, runway-tts, runway-sts, or runway-dubbing, or asks about Runway in Nebula. Sourced from the official Runway Python SDK (github.com/runwayml/sdk-python) cross-checked against backend/handlers/runway.py and backend/data/node_definitions.json, plus the Nebula audit guide (docs/api-guides/runway.md), on 2026-06-04 (updated 2026-06-10).

OpenRouter universal gateway in Nebula — one node (openrouter-universal) that reaches hundreds of models (GPT, Claude, Gemini, Llama, DeepSeek, FLUX, Recraft, and more) through OpenRouter's chat-completions API. Capabilities exposed in Nebula: streaming text/chat/reasoning/coding, vision (image-in → text-out), and image generation (text → image) — the model you pick decides which. Activate when the user configures the `openrouter-universal` node (display name "OpenRouter"), picks/searches an OpenRouter model in the Inspector, or asks about OpenRouter in Nebula. Sourced from the official OpenRouter docs (openrouter.ai/docs) and the Nebula audit guide docs/api-guides/openrouter.md on 2026-06-04 — every node id, port, param, header, and capability boundary here is cross-checked against backend/handlers/openrouter.py, backend/routes/openrouter_proxy.py, and backend/data/node_definitions.json.

Krea API in Nebula — generate images with Krea's own Krea 2 model steered by reference images, moodboards, and custom-trained LoRA styles, plus search and train Krea styles. Activate when the user configures any krea-2-generate, krea-style-search, or krea-style-train node, or asks about Krea in Nebula. Sourced from docs.krea.ai and cross-checked against backend/handlers/krea.py + backend/data/node_definitions.json on 2026-06-04; see the Nebula audit guide at docs/api-guides/krea.md.

Meshy 3D generation API — text-to-3D, image-to-3D, multi-image-to-3D, remesh, retexture, auto-rigging, animation, 3D printing, plus nano-banana-backed 2D gen. Activate when the user asks for 3D models, rigging, animations, 3D printing, or configures any node whose id starts with `meshy-`. Sourced directly from docs.meshy.ai on 2026-04-17 — every parameter, endpoint, and price quoted here came from the canonical API reference.

Use when building or editing a Nebula graph containing Krea 2 direct API nodes, Krea image style references, Krea styles, moodboards, style search, or style training. Covers node IDs, graph wiring, params, provider resource objects, direct-Krea-only routing, API-key requirements, and live-test caveats.

Use when building or editing a Nebula graph containing a gpt-image-2-* node (generate, edit, fal-generate, fal-edit) or when the user asks to use gpt-image-2 inside Nebula. Covers Nebula-specific node IDs, param names, and UI wiring.

Meshy 3D generation API — text-to-3D, image-to-3D, multi-image-to-3D, remesh, retexture, auto-rigging, animation, 3D printing, plus nano-banana-backed 2D gen. Activate when the user asks for 3D models, rigging, animations, 3D printing, or configures any node whose id starts with `meshy-`. Sourced directly from docs.meshy.ai on 2026-04-17 — every parameter, endpoint, and price quoted here came from the canonical API reference.

Runway API — Gen-4, Gen-4 Image, Gen-4 Aleph, Act-Two character performance, Veo 3/3.1 pass-through, Seedance 2, Eleven-Labs-backed TTS/STS/dubbing. Activate when the user configures any `runway-*` node or asks about Runway's API. Sourced from the official Stainless-generated Python SDK (github.com/runwayml/sdk-python) on 2026-04-17.

Daedalus's iterative-craftsman playbook — pipeline-stage tracing + vision reliability rules + nebula CLI cookbook + learnings discipline + autonomy modes. Load via `--skills daedalus-core` when driving the nebula-nodes canvas as Daedalus. (Persona/identity lives in the profile's SOUL.md, not here.)

Use when building or editing a Nebula graph containing a gpt-image-2-* node (generate, edit, fal-generate, fal-edit) or when the user asks to use gpt-image-2 inside Nebula. Covers Nebula-specific node IDs, param names, and UI wiring. For prompting craft, see the global gpt-image-2 skill.