Source: Higgsfield Mcp 2026 04 28 (Higgsfield MCP product page — https://higgsfield.ai/mcp) 2026-10-27 addition (Katana): raw/Claude_Opus_5.5_Made_Clipping_The_Easiest_Side_Hustle_of_2026.md (Higgsfield’s own creator; first-party marketing), raw/Anthropic_Just_Dropped_Claude_Motion_Dashboards_+_More_AI_News.md, raw/ai_put_minecraft_in_skyrim_and_open_ai_had_a_dev_day.md. 2026-10-30 addition (ChatGPT plugin, Seedance 2.5 motion graphics): raw/GPT-6_Astra_+_Higgsfield_AI_-_How_to_Make_AI_Videos_in_ChatGPT_Beginner_s_Guide.md, raw/Claude_+_Higgsfield_-_How_to_Make_Motion_Graphics_with_AI_Full_AI_Workflow.md (AI Master). 2026-10-30 addition (Hermes setup, credit costs): raw/Hermes_Agent_Can_Make_Videos_Now_This_Changes_Everything.md (Sharbel, sponsored).

Higgsfield ships an MCP server that drops their full image-and-video generation stack into Claude (or any MCP-compatible agent) as a custom connector. It’s the conversational, web-first counterpart to the Python SDK / REST API — no API keys, no polling code, no request_id bookkeeping. You add one connector, sign in, and Claude can generate, browse, compare, and iterate on cinematic stills and clips inside the chat.

Key Takeaways

  • Connector URL: https://mcp.higgsfield.ai/mcp. Add it as a custom connector in Claude’s settings (Settings → Connectors → Add custom connector → name “Higgsfield” → paste URL → Connect → sign in with Higgsfield account).
  • No API keys. Auth flows through your existing Higgsfield account on connect — different from the SDK path, which requires HF_KEY or HF_API_KEY+HF_API_SECRET env vars.
  • Supported clients: Claude (web, desktop, Claude Code), OpenClaw, Hermes, NemoClaw — and any MCP-compatible client.^[inferred] The page enumerates these four by name.
  • Image generation: 16+ models from a single connection. Models named on the page: Soul, Nano Banana Pro, Flux, Seedream. Up to 4K output, configurable aspect ratio.
  • Video generation: 17+ models. Models named on the page: Seedance, Kling, Veo, Minimax Hailuo. Controls for duration, aspect ratio, genre, and start/end frames.
  • Asset library is queryable. Claude can search prior generations, display specific results, and use any previous output as a reference for a new generation — Higgsfield’s history becomes part of the agent’s working memory.
  • Product Ads & Videos preset. Drop a URL or upload a photo and the connector returns campaign-ready images and videos. One prompt fans out to multiple formats.
  • 9 curated video presets: UGC, unboxing, product review, hyper motion, TV spot, and more. The agent passes the chosen mode into generation automatically.
  • Soul Characters for cast consistency. Train a character from reference photos in one turn, then generate scenes/lookbooks/videos around that character across the rest of the conversation.
  • Multi-model comparison. “Generate this scene on 4 different models and show me the results” — same prompt, different models, side by side. Pick the winner and iterate.
  • Credits, not API pricing. The MCP rides on your existing Higgsfield plan credits — no separate billing structure.^[inferred from the page’s promise that it works with your existing Higgsfield account; the page itself doesn’t break down credit costs per model.]

Setup (3 steps)

  1. Open Claude → Settings → Connectors.
  2. Add a custom connector, name it “Higgsfield”, and paste the URL https://mcp.higgsfield.ai/mcp.
  3. Connect and sign in with your existing Higgsfield account.

That’s it — no .mcp.json, no environment variables, no polling loop. From the next message onward Claude can call into image generation, video generation, the asset library, and the marketing-studio presets.

What changes vs the API/SDK path

SurfaceAuthProgrammabilityBest for
MCP connector (this article)Sign in once, no keysConversational — Claude picks model + paramsDay-to-day creative work, product ads, character-driven campaigns, single-prompt workflows
**[[ai-video-content/higgsfield-sdkPython SDK]]**HF_KEY or key+secret env varsCode — submit / poll / callback / managed request patterns
**[[ai-video-content/higgsfield-overviewDirect REST]]**API key + secretRaw HTTPS to platform.higgsfield.ai

Same models, same credit pool, different ergonomics. The MCP is for the agent surface; the SDK is for engineering surfaces. They don’t conflict — connect both if you do both.

Use Cases (per the page)

  • E-Commerce & Product — lifestyle product shots, background swaps, promotional videos.^[inferred] The page lists “E-Commerce & Product” as a use-case card; the elaboration is consistent with the Product Ads preset description but isn’t quoted verbatim.
  • Social Media & Content Creation — scroll-optimized images and short-form videos sized for the major channels.
  • Marketing Agencies — multi-format campaign variations and client-ready assets.
  • Filmmaking — storyboarding, concept art, previsualization, consistent character casting (Soul Characters).
  • Infographics & Visual Data — custom illustrations and supporting imagery.

Three workflow shapes

The page calls out three paths in increasing depth:

  1. Generate one asset in seconds. Describe what you need; Claude picks the model, sets parameters, delivers. Example prompt the page shows: “Generate a cinematic wide shot of a neon-lit Tokyo alley at night.”
  2. Build a visual system from a conversation. Train a Soul Character, generate scenes across locations and styles, produce videos, manage everything from your generation history. Example prompt: “Train a Soul Character from these photos, then generate a 10-image lookbook.”
  3. Compare models side by side. Same prompt across Flux, Soul, Cinema Studio, and Seedream. Pick the winner and iterate. Example prompt: “Generate this scene on 4 different models and show me the results.”

Implementation

  • Tool/Service: Higgsfield MCP server (https://mcp.higgsfield.ai/mcp) — product page at https://higgsfield.ai/mcp.
  • Setup: Three steps above. No API keys; OAuth-style sign-in on connect.^[inferred — the page describes “Connect and sign in” but doesn’t name the auth protocol.]
  • Cost: Existing Higgsfield credits. The page does not publish per-model credit costs; check the Higgsfield Cloud dashboard.
  • Integration notes:
    • Treat the MCP as the conversational surface and reach for the SDK when you need scheduled, headless, or webhook-driven generation.
    • Soul Characters live in the same account — character trained via the MCP shows up in the SDK and vice versa.^[inferred from the shared-account model.]
    • The “Product Ads” preset accepts a URL or an uploaded photo. For DTC clients, this is the lowest-friction entry — drop the product page URL into the conversation.
    • Multi-model side-by-side is a credit multiplier: 4 models × 1 prompt = 4 generations billed.

Positioning vs other AI video tooling

  • vs the REST API + SDK / webhooks surfaces — same backend, different consumer. MCP for humans-in-conversation; SDK for systems.
  • vs HeyGen Avatar V — HeyGen is talking-head-first (avatars, lip-sync, presenters); Higgsfield MCP is generative-image/video-first (scenes, products, characters across stills + clips). Different jobs to be done, often complementary in a campaign.
  • vs HeyGen Studio Automation with Claude Code — that pattern uses Claude Code to script HeyGen jobs. Higgsfield MCP collapses the scripting layer: you talk to Claude and Claude calls Higgsfield directly.
  • vs Remotion / Hyperframes — those are deterministic composition layers (React + HTML). Higgsfield generates the raw clips and stills the composition layer arranges. Common pipeline: Higgsfield MCP → save best generations to disk → Remotion or Hyperframes for sequencing, captions, brand frame.
  • vs video-use — video-use is a Claude Code editing skill (transcripts, cuts, animation sub-agents). It edits footage you already have. Higgsfield MCP creates the footage to feed it.

October 2026 update: Katana, edit-to-the-beat inside Claude

Higgsfield added Katana, an editing feature run through this MCP. Two sources cover it, and both are promotional. The clipping video is by Higgsfield’s own creator, Adil (“I’ve already tried it with our own movie”), so treat it as first-party marketing. The second source is a creator’s news roundup (raw/Anthropic_Just_Dropped_Claude_Motion_Dashboards_+_More_AI_News.md).

  • What it is. “Higgsfield says it’s powered by Claude Motion” (Claude Motion) and it runs inside Claude through the Higgsfield MCP. Connect the MCP as above, then type /katana in the chat.
  • Recipe (Adil). Select Opus 5.5, connect Katana through the MCP, and add Higgsfield’s “clipping skill” (linked from the video description, not captured). Give Claude the footage or a link, a reference edit and the music. In the prompt, set the length, ask for vertical format, a strong opening, pacing that matches the reference and cuts synced to the music, and “a quick report after each stage”. Per Adil, Katana picks the strongest moments and builds the edit, while Claude works out the characters and plot. He reports that the result recreated the reference style “almost one to one” with cuts on the beat.
  • Pick the reference on evidence. “A bad reference can mean a bad result.” Check its views, likes and shares, and read the comments to see what viewers react to.
  • Other formats shown. A podcast clip built around one strong statement, with captions and generated graphics. A 30-second teaser built on “what happens next” instead of “this guy’s cool”; ask Claude for several teaser angles from one film. Translated versions of the same clips as an upsell.
  • Speed and quality (news-roundup creator). One Katana generation from a reference video took “about like 20 minutes”. The creator judged it “much better than any of the outputs that I’ve gotten so far” from plain Claude Motion. That is opinion from one run.
  • Clipping economics (Adil’s claims, unsourced). Campaign platforms advertise “50 cents per 1,000 eligible views”. A sample podcast package is “10 clips for 3 billion” clipping economy and the “1.4 million in just 5 weeks” streamer spend are stated without sources. The full economics breakdown belongs in the Adil content-factory article.
  • Aggregator MCPs for long-form agent video. AI For Humans name Higgsfield alongside Runway and Magnific as MCP-bearing platforms an agent can drive for a full documentary. See Agent-Made AI Documentary.

October 2026 update: ChatGPT plugin route and Seedance 2.5 motion graphics (AI Master)

Two AI Master videos (fetched 2026-10-30) add a client and a model to this article. Higgsfield is linked in both descriptions and is the only tool shown; neither transcript says whether the video is sponsored.

  • ChatGPT is now a supported client, through a plugin. The product page reviewed in April named Claude, OpenClaw, Hermes and NemoClaw. Higgsfield’s MCP section now lists ChatGPT as a client with an Add Higgsfield plugin button that opens ChatGPT’s install screen; install, sign in, then Start creating loads a test prompt that generates an image through Nano Banana. Desktop only for the install. This is an addition to the client list, not a conflict. Walkthrough with GPT-6 Astra: Higgsfield in ChatGPT with GPT-6 Astra.
  • Claude setup is unchanged: copy the connector URL from Higgsfield’s MCP and CLI section, then Claude → Settings → Connectors → Add custom connector.
  • Seedance 2.5 renders motion graphics from Claude. Six builds (title card, reference-styled product moment, product light sweep, title sequence, burger menu, 30-second explainer) with legible text and objects that held their shape, per the presenter. Details and his four prompting habits: Claude + Higgsfield Motion Graphics.

Hermes setup and per-job credit costs (Sharbel, August 2026)

Source: raw/Hermes_Agent_Can_Make_Videos_Now_This_Changes_Everything.md (Sharbel, YouTube @sharbelxyz, https://www.youtube.com/watch?v=mMUS-BCHyc8, 13:28, uploaded 2026-08-27). Sponsored by Higgsfield, which supplied the credits. Captions spell the product “Hicksfield” and the model “Seedens”/“Seasons”; read here as Higgsfield and Seedance.

  • One URL for every client. He shows the same endpoint (mcp.higgsfield.ai/mcp) listed for Claude, Grok Bot, ChatGPT, Cursor, Claude Code, OpenClaw and Hermes. This answers the Open Question below on whether non-Claude clients need their own setup.
  • Auth. In Claude: plus, Connectors, Add custom connector, paste the URL, accept the recommended options, then an allow-access window. In Hermes, which is MCP-native: paste the Hermes snippet from Higgsfield’s page into the chat; Hermes returns a link to authenticate, then selects the workspace.
  • The agent picks the model. For a product-shot prompt it chose a Seedance model on its own.
  • Credit costs he reports (his conversion to dollars):
JobCreditsAbout
One product-shot video (matte black bottle, hard side light), under a minute45$2.50
15-second UGC ad from that bottle, with a talking presenter, about 4 minutes75$4
10-second 3D cartoon YouTube intro: concept, four stills, four shots136$7.50
Responsive hero website with scroll animation, about 5 minutes41$2
  • His notes on output. The UGC voice sounded AI-generated; the intro invented a Higgsfield logo, so give it real brand assets (he suggests a brand-asset folder). He asked the agent to help write the prompt and ask him questions before each run.
  • Why Hermes rather than Claude. Turn a prompt into a skill holding brand colours, style and output folder, then attach it to a cron: for each new video, read the title and transcript and drop three intro options in the editor’s folder.

This partly answers the per-model credit-cost Open Question below, for four of his jobs. It does not give per-model rates.

November 2026 update: talking-head setting swaps from Claude Code (Brock Mesarich)

Source: raw/Fable_5.2_Is_INCREDIBLE_at_Video_Editing_WILD_RESULTS.md (Brock Mesarich, https://www.youtube.com/watch?v=1fBhLGhQQ2M).

  • Brock’s Claude Code editing system calls Higgsfield to re-shoot a talking-head clip in a different outfit or room (a gray hoodie in the same room, a denim jacket in a brick loft); he had Claude make 11 versions of one clip.
  • The swaps used Kling inside Higgsfield. In his words, “each one of these swaps cost 12 different credits, and it took between 4 and 8 minutes” per video.
  • Full system: Brock Mesarich’s Claude Code video-editing system.

Open Questions

  • Per-model credit costs. The page promises “your existing plan credits” but does not list cost per Soul / Flux / Seedream / Seedance / Kling / Veo / Minimax Hailuo generation.
  • Rate limits inside MCP. Same pool as REST? Different? Not stated.
  • Tool names exposed by the MCP. The page describes capabilities (image generation, video generation, asset browsing, product ads, presets) but does not enumerate the exact tool names the MCP advertises to clients.
  • Auth protocol. “Connect and sign in” is shown as a step but the page doesn’t specify OAuth vs custom. Worth verifying when wiring into Hermes or another non-Claude client.
  • Resolution/duration ceilings per model. “Up to 4K” for images and “control duration” for video — explicit per-model max not published here.
  • Hermes/OpenClaw/NemoClaw integration shape. Partly resolved 2026-10-30: Sharbel’s video shows one URL for all clients and a paste-snippet-then-authenticate flow in Hermes (see § Hermes setup). NemoClaw was not shown.

Try It

  1. Open Claude → Settings → Connectors → Add custom connector. Name it “Higgsfield”, paste https://mcp.higgsfield.ai/mcp, click Add → Connect → sign in. Total time: ~30 seconds.
  2. Smoke-test with a single image. Try the page’s example prompt verbatim: “Generate a cinematic wide shot of a neon-lit Tokyo alley at night.” This is the cheapest way to confirm the connector is alive and credits flow.
  3. Run the multi-model comparison. “Generate this scene on 4 different models and show me the results.” Worth the credit burn once — you’ll calibrate which model is your default for which job.
  4. Train a Soul Character on a real client. Pick a recurring face/product and run the lookbook prompt: “Train a Soul Character from these photos, then generate a 10-image lookbook.” Soul Characters carry across generations, so the cost is one-time training + cheap reuses.
  5. For a Smile Springs–style dental client: drop the practice’s homepage URL into the Product Ads preset and ask for a 9-format campaign sweep (UGC, unboxing-of-new-patient-welcome-kit, hyper motion, TV spot). One prompt → multi-channel pack — much closer to a marketer’s day than the SDK path.
  6. If you outgrow the conversational surface, graduate the workflow to the SDK + webhooks for headless / scheduled / pipeline integration. The MCP and SDK share the same account and credit pool, so the migration is no-drama.