Skip to main content
Glama

Avocado AI

by co.avocadoai.mcp
OverviewDiscussionsAdmin

Server Details

Create ads inside any AI assistant with Avocado, create, edit and make AI UGC in chat.

Glama couldn't complete the latest health check. If this server requires authentication, missing or expired test credentials may be the cause. A test profile lets Glama authenticate for health checks and discover tools; it is separate from your personal connections.

If you are the author, claim ownership, then add or update a test profile under Admin → Test Profile.

Status
Unhealthy
Uptime
0.0% over 37 days
OAuth
Works in Glama
Last Tested
Transport
Streamable HTTP
URL

TDQS

A4.1/5.0

Scored across 17 tools

Disambiguation5/5

Each tool targets a distinct media type or action (credit check, job polling, flow creation, storyboard management, various media generation, upload preparation, model listing, onboarding). No two tools have overlapping purposes, and even the paired 'generate to storyboard' tools are clearly distinct from their base generation counterparts.

Naming Consistency4/5

Most tools follow a verb_noun pattern (e.g., generate_image, list_storyboards, create_flow). However, 'models_list' deviates from the pattern (should be 'list_models'), and 'get_started' uses a different verb tense. Overall, the convention is mostly consistent.

Tool Count4/5

17 tools is on the higher side but justified by the server's broad scope (image, video, music, speech, effects, flow, storyboard, account, models, upload). Each tool has a clear role, though a few (like 'generate_image_to_storyboard') could be optional parameters rather than separate tools.

Completeness3/5

The tool surface covers core generation tasks (image, video, music, speech, effects) and supporting operations (upload, polling, account, models). However, storyboard management is missing update/delete functionality, and there is no tool for cancelling jobs or editing generated videos/audio.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

  1. 3 tool updates
    • Addedcancel_scheduled_job
    • Addedlist_scheduled_jobs
    • Addedschedule_job
  2. 1 tool update
    • Changedgenerate_image_to_flow1 field changed
      • changedInput schema / properties / reference_image_urls / description
        Previous value: -"HTTPS URLs of already-generated reference images to KEEP CONSISTENT — pass the cast (character) image URL(s) and/or the location image URL to steer generation. Must be permanent URLs (Supabase or fal-hosted) — it doesn't matter which tool produced them, pasted/temporary URLs are dropped and you'll be told in the response. When set, the image is built with an edit/reference model (nano-banana-2 by default, or gpt-image-2) instead of plain text-to-image. For role 'beat' you can normally OMIT this — see the tool description."New value: +"HTTPS URLs of already-generated reference images to KEEP CONSISTENT — pass the character sheet URL of EACH character appearing in the shot and/or the location image URL to steer generation. Must be permanent URLs (Supabase or fal-hosted) — it doesn't matter which tool produced them, pasted/temporary URLs are dropped and you'll be told in the response. When set, the image is built with an edit/reference model (nano-banana-2 by default, or gpt-image-2) instead of plain text-to-image. For role 'beat' you can normally OMIT this — see the tool description."
  3. 3 tool updates
    • Changedpublish_post3 fields changed
      • addedInput schema / properties / document_title
        Added value: +{
+  "description": "LinkedIn document posts only: the title shown on the document carousel. Defaults to the file name if omitted.",
+  "type": "string"
+}
      • addedInput schema / properties / first_comment
        Added value: +{
+  "description": "LinkedIn only: text auto-posted as the first comment. Put external links here instead of the caption for better reach.",
+  "type": "string"
+}
      • changedInput schema / properties / media_urls / description
        Previous value: -"Public media URLs to attach (the asset URLs you generated). Required for Instagram/TikTok."New value: +"Public media URLs to attach (the asset URLs you generated). Required for Instagram/TikTok. For a LinkedIn document post, exactly one PDF/PPT/PPTX/DOC/DOCX direct file URL."
    • Addedread_document
    • Changedschedule_post3 fields changed
      • addedInput schema / properties / document_title
        Added value: +{
+  "description": "LinkedIn document posts only: the title shown on the document carousel. Defaults to the file name if omitted.",
+  "type": "string"
+}
      • addedInput schema / properties / first_comment
        Added value: +{
+  "description": "LinkedIn only: text auto-posted as the first comment. Put external links here instead of the caption for better reach.",
+  "type": "string"
+}
      • changedInput schema / properties / media_urls / description
        Previous value: -"Public media URLs to attach. Required for Instagram/TikTok."New value: +"Public media URLs to attach. Required for Instagram/TikTok. For a LinkedIn document post, exactly one PDF/PPT/PPTX/DOC/DOCX direct file URL."
  4. 2 tool updates
    • Changedpublish_post3 fields changed
      • changedInput schema / properties / content / description
        Previous value: -"The post caption / text. Keep it platform-appropriate."New value: +"The post caption / text. Keep it platform-appropriate (X max 280 chars, Threads max 500)."
      • changedInput schema / properties / platforms / description
        Previous value: -"Which connected platforms to post to. Use list_connected_accounts to see what's linked."New value: +"Which connected platforms to post to ('twitter' = X). Use list_connected_accounts to see what's linked."
      • changedInput schema / properties / platforms / items / enum
        Previous value: -[
-  "instagram",
-  "tiktok",
-  "linkedin"
-]New value: +[
+  "instagram",
+  "tiktok",
+  "linkedin",
+  "twitter",
+  "threads"
+]
    • Changedschedule_post3 fields changed
      • changedInput schema / properties / content / description
        Previous value: -"The post caption / text."New value: +"The post caption / text (X max 280 chars, Threads max 500)."
      • changedInput schema / properties / platforms / description
        Previous value: -"Which connected platforms to post to."New value: +"Which connected platforms to post to ('twitter' = X)."
      • changedInput schema / properties / platforms / items / enum
        Previous value: -[
-  "instagram",
-  "tiktok",
-  "linkedin"
-]New value: +[
+  "instagram",
+  "tiktok",
+  "linkedin",
+  "twitter",
+  "threads"
+]
  5. 4 tool updates
    • Changedgenerate_speech_to_flow2 fields changed
      • changedInput schema / properties / voice / description
        Previous value: -"Voice to use. Defaults to 'rachel'."New value: +"Voice to use. Defaults to 'rachel' (ElevenLabs). vivi/mindy/kian/sophie/magnus/nadia are Seed Audio 1.0 (multilingual, max 2048 chars)."
      • changedInput schema / properties / voice / enum
        Previous value: -[
-  "rachel",
-  "adam",
-  "antoni",
-  "bella",
-  "domi",
-  "elli",
-  "josh",
-  "arnold",
-  "sam"
-]New value: +[
+  "rachel",
+  "adam",
+  "antoni",
+  "bella",
+  "domi",
+  "elli",
+  "josh",
+  "arnold",
+  "sam",
+  "vivi",
+  "mindy",
+  "kian",
+  "sophie",
+  "magnus",
+  "nadia"
+]
    • Changedgenerate_video2 fields changed
      • changedInput schema / properties / model / enum
        Previous value: -[
-  "seedance-2.0-t2v",
-  "seedance-2.0-t2v-fast",
-  "seedance-2.0-i2v",
-  "seedance-2.0-i2v-fast",
-  "seedance-2.0-ref",
-  "seedance-2.0-ref-fast",
-  "seedance-2.0-t2v-mini",
-  "seedance-2.0-i2v-mini",
-  "seedance-2.0-ref-mini",
-  "kling3-standard",
-  "kling3-pro",
-  "kling3-4k",
-  "kling-o3-4k",
-  "grok-imagine-video-v1-5",
-  "happy-horse-t2v",
-  "happy-horse-i2v"
-]New value: +[
+  "seedance-2.0-t2v",
+  "seedance-2.0-t2v-fast",
+  "seedance-2.0-i2v",
+  "seedance-2.0-i2v-fast",
+  "seedance-2.0-ref",
+  "seedance-2.0-ref-fast",
+  "seedance-2.0-t2v-mini",
+  "seedance-2.0-i2v-mini",
+  "seedance-2.0-ref-mini",
+  "kling3-standard",
+  "kling3-pro",
+  "kling3-4k",
+  "kling-o3-4k",
+  "grok-imagine-video-v1-5",
+  "happy-horse-t2v",
+  "happy-horse-i2v",
+  "gemini-omni-flash",
+  "gemini-omni-flash-ref",
+  "gemini-omni-flash-edit"
+]
      • addedInput schema / properties / video_url
        Added value: +{
+  "description": "gemini-omni-flash-edit only: HTTPS URL of the source video to edit (e.g. a previously generated video's URL). Required for that model; ignored by all others.",
+  "type": "string"
+}
    • Changedgenerate_video_to_flow1 field changed
      • changedInput schema / properties / model / enum
        Previous value: -[
-  "seedance-2.0-t2v",
-  "seedance-2.0-t2v-fast",
-  "seedance-2.0-i2v",
-  "seedance-2.0-i2v-fast",
-  "seedance-2.0-ref",
-  "seedance-2.0-ref-fast",
-  "seedance-2.0-t2v-mini",
-  "seedance-2.0-i2v-mini",
-  "seedance-2.0-ref-mini",
-  "kling3-standard",
-  "kling3-pro",
-  "kling3-4k",
-  "kling-o3-4k",
-  "happy-horse-t2v",
-  "happy-horse-i2v"
-]New value: +[
+  "seedance-2.0-t2v",
+  "seedance-2.0-t2v-fast",
+  "seedance-2.0-i2v",
+  "seedance-2.0-i2v-fast",
+  "seedance-2.0-ref",
+  "seedance-2.0-ref-fast",
+  "seedance-2.0-t2v-mini",
+  "seedance-2.0-i2v-mini",
+  "seedance-2.0-ref-mini",
+  "kling3-standard",
+  "kling3-pro",
+  "kling3-4k",
+  "kling-o3-4k",
+  "happy-horse-t2v",
+  "happy-horse-i2v",
+  "gemini-omni-flash"
+]
    • Changedgenerate_video_to_storyboard1 field changed
      • changedInput schema / properties / model / enum
        Previous value: -[
-  "seedance-2.0-t2v",
-  "seedance-2.0-t2v-fast",
-  "seedance-2.0-i2v",
-  "seedance-2.0-i2v-fast",
-  "seedance-2.0-ref",
-  "seedance-2.0-ref-fast",
-  "seedance-2.0-t2v-mini",
-  "seedance-2.0-i2v-mini",
-  "seedance-2.0-ref-mini",
-  "kling3-standard",
-  "kling3-pro",
-  "kling3-4k",
-  "kling-o3-4k",
-  "happy-horse-t2v",
-  "happy-horse-i2v"
-]New value: +[
+  "seedance-2.0-t2v",
+  "seedance-2.0-t2v-fast",
+  "seedance-2.0-i2v",
+  "seedance-2.0-i2v-fast",
+  "seedance-2.0-ref",
+  "seedance-2.0-ref-fast",
+  "seedance-2.0-t2v-mini",
+  "seedance-2.0-i2v-mini",
+  "seedance-2.0-ref-mini",
+  "kling3-standard",
+  "kling3-pro",
+  "kling3-4k",
+  "kling-o3-4k",
+  "happy-horse-t2v",
+  "happy-horse-i2v",
+  "gemini-omni-flash"
+]
  6. 4 tool updates
    • Changededit_image2 fields changed
      • changedInput schema / properties / model / description
        Previous value: -"Edit model. 'nano-banana-2' is fast and cheap (default). 'gpt-image-2' is higher quality but costs more credits."New value: +"Edit model. 'nano-banana-2' is fast and cheap (default). 'nano-banana-2-lite' is even faster/cheaper for simple single-image touch-ups (not optimized for multiple reference images). 'gpt-image-2' is higher quality but costs more credits."
      • changedInput schema / properties / model / enum
        Previous value: -[
-  "nano-banana-2",
-  "gpt-image-2"
-]New value: +[
+  "nano-banana-2",
+  "nano-banana-2-lite",
+  "gpt-image-2"
+]
    • Changededit_image_to_flow7 fields changed
      • changedInput schema / properties / beat_number / description
        Previous value: -"For role 'beat': the beat's position in story order (1..N)."New value: +"For role 'beat': the beat's position in story order (1..N). With replace_node_id, defaults to the existing tile's current beat_number."
      • changedInput schema / properties / label / description
        Previous value: -"Short human label shown in the Director Library, e.g. 'Pip - hero (v2)' or 'Beat 3 - Into the Backrooms'."New value: +"Short human label shown in the Director Library, e.g. 'Pip - hero (v2)' or 'Beat 3 - Into the Backrooms'. With replace_node_id, defaults to the existing tile's current label."
      • changedInput schema / properties / model / description
        Previous value: -"Edit model. 'nano-banana-2' is fast and cheap (default). 'gpt-image-2' is higher quality but costs more."New value: +"Edit model. 'nano-banana-2' is fast and cheap (default). 'nano-banana-2-lite' is even faster/cheaper for simple single-image touch-ups (not optimized for multiple reference images). 'gpt-image-2' is higher quality but costs more."
      • changedInput schema / properties / model / enum
        Previous value: -[
-  "nano-banana-2",
-  "gpt-image-2"
-]New value: +[
+  "nano-banana-2",
+  "nano-banana-2-lite",
+  "gpt-image-2"
+]
      • changedInput schema / properties / num_images / description
        Previous value: -"Number of edited variants (1-4). Defaults to 1. One tile per variant."New value: +"Number of edited variants (1-4). Defaults to 1. One tile per variant. Not compatible with replace_node_id (which always produces exactly one)."
      • addedInput schema / properties / replace_node_id
        Added value: +{
+  "description": "Regenerate an EXISTING tile in place instead of creating a new one — use when the user asks to redo/regenerate/tweak a specific cast, location, or beat tile that's already on the flow. The node keeps its position, and (unless overridden) its role/label/beat_number. Get the id from a prior generate_image_to_flow/edit_image_to_flow response's nodeIds, or from list_flow_assets.",
+  "type": "string"
+}
      • changedInput schema / properties / role / description
        Previous value: -"Director Library grouping. 'cast' = a character reference, 'location' = a place/world reference, 'beat' = a storyboard frame. Omit for a plain edited image."New value: +"Director Library grouping. 'cast' = a character reference, 'location' = a place/world reference, 'beat' = a storyboard frame. Omit for a plain edited image. With replace_node_id, defaults to the existing tile's current role."
    • Changedgenerate_image_to_flow6 fields changed
      • changedInput schema / properties / beat_number / description
        Previous value: -"For role 'beat': the beat's position in story order (1..N). Keeps frames and clips ordered in the Library."New value: +"For role 'beat': the beat's position in story order (1..N). Keeps frames and clips ordered in the Library. With replace_node_id, defaults to the existing tile's current beat_number."
      • changedInput schema / properties / label / description
        Previous value: -"Short human label shown in the Director Library, e.g. 'Pip - hero character' or 'Beat 3 - Into the Backrooms'."New value: +"Short human label shown in the Director Library, e.g. 'Pip - hero character' or 'Beat 3 - Into the Backrooms'. With replace_node_id, defaults to the existing tile's current label."
      • changedInput schema / properties / num_images / description
        Previous value: -"Number of images to generate (1-4). Defaults to 1. One tile per image."New value: +"Number of images to generate (1-4). Defaults to 1. One tile per image. Not compatible with replace_node_id (which always produces exactly one)."
      • changedInput schema / properties / reference_image_urls / description
        Previous value: -"HTTPS URLs of already-generated reference images to KEEP CONSISTENT — pass the cast (character) image URL(s) and the location image URL when generating a beat frame, so the same character and setting carry across every beat. When set, the image is built with an edit/reference model (nano-banana-2 by default, or gpt-image-2) instead of plain text-to-image. Use the URLs returned by your earlier cast/location generations."New value: +"HTTPS URLs of already-generated reference images to KEEP CONSISTENT — pass the cast (character) image URL(s) and/or the location image URL to steer generation. Must be permanent URLs (Supabase or fal-hosted) — it doesn't matter which tool produced them, pasted/temporary URLs are dropped and you'll be told in the response. When set, the image is built with an edit/reference model (nano-banana-2 by default, or gpt-image-2) instead of plain text-to-image. For role 'beat' you can normally OMIT this — see the tool description."
      • addedInput schema / properties / replace_node_id
        Added value: +{
+  "description": "Regenerate an EXISTING tile in place instead of creating a new one — use when the user asks to redo/regenerate/iterate on a specific cast, location, or beat tile that's already on the flow. The node keeps its position, and (unless overridden) its role/label/beat_number. Get the id from a prior generate_image_to_flow/edit_image_to_flow response's nodeIds, or from list_flow_assets.",
+  "type": "string"
+}
      • changedInput schema / properties / role / description
        Previous value: -"Director Library grouping. 'cast' = a character reference, 'location' = a place/world reference, 'beat' = a single storyboard frame (still) for a beat. Omit for a plain generated image."New value: +"Director Library grouping. 'cast' = a character reference, 'location' = a place/world reference, 'beat' = a single storyboard frame (still) for a beat. Omit for a plain generated image. With replace_node_id, defaults to the existing tile's current role."
    • Addedlist_flow_assets
  7. 2 tool updates
    • Addededit_image_to_flow
    • Changedgenerate_image_to_flow1 field changed
      • addedInput schema / properties / reference_image_urls
        Added value: +{
+  "description": "HTTPS URLs of already-generated reference images to KEEP CONSISTENT — pass the cast (character) image URL(s) and the location image URL when generating a beat frame, so the same character and setting carry across every beat. When set, the image is built with an edit/reference model (nano-banana-2 by default, or gpt-image-2) instead of plain text-to-image. Use the URLs returned by your earlier cast/location generations.",
+  "items": {
+    "type": "string"
+  },
+  "maxItems": 6,
+  "type": "array"
+}
  8. 1 tool update
    • Changedgenerate_video2 fields changed
      • changedInput schema / properties / resolution / description
        Previous value: -"Video resolution. Meaningful for seedance (480p/720p/1080p; 1080p not allowed with seedance fast/ref-fast) and happy-horse (720p/1080p, default 720p). Kling models lock resolution by variant."New value: +"Video resolution. Meaningful for seedance Standard (480p/720p/1080p/4k; Fast & Mini cap at 480p/720p — no 1080p/4k) and happy-horse (720p/1080p, default 720p). Seedance 4k is Growth/Pro only. Kling models lock resolution by variant."
      • changedInput schema / properties / resolution / enum
        Previous value: -[
-  "480p",
-  "720p",
-  "1080p",
-  "4K"
-]New value: +[
+  "480p",
+  "720p",
+  "1080p",
+  "4K",
+  "4k"
+]
  9. 2 tool updates
    • Addedgenerate_music_to_flow
    • Addedgenerate_speech_to_flow
  10. 3 tool updates
    • Changedgenerate_video1 field changed
      • changedInput schema / properties / model / enum
        Previous value: -[
-  "seedance-2.0-t2v",
-  "seedance-2.0-t2v-fast",
-  "seedance-2.0-i2v",
-  "seedance-2.0-i2v-fast",
-  "seedance-2.0-ref",
-  "seedance-2.0-ref-fast",
-  "kling3-standard",
-  "kling3-pro",
-  "kling3-4k",
-  "kling-o3-4k",
-  "grok-imagine-video-v1-5",
-  "happy-horse-t2v",
-  "happy-horse-i2v"
-]New value: +[
+  "seedance-2.0-t2v",
+  "seedance-2.0-t2v-fast",
+  "seedance-2.0-i2v",
+  "seedance-2.0-i2v-fast",
+  "seedance-2.0-ref",
+  "seedance-2.0-ref-fast",
+  "seedance-2.0-t2v-mini",
+  "seedance-2.0-i2v-mini",
+  "seedance-2.0-ref-mini",
+  "kling3-standard",
+  "kling3-pro",
+  "kling3-4k",
+  "kling-o3-4k",
+  "grok-imagine-video-v1-5",
+  "happy-horse-t2v",
+  "happy-horse-i2v"
+]
    • Changedgenerate_video_to_flow1 field changed
      • changedInput schema / properties / model / enum
        Previous value: -[
-  "seedance-2.0-t2v",
-  "seedance-2.0-t2v-fast",
-  "seedance-2.0-i2v",
-  "seedance-2.0-i2v-fast",
-  "seedance-2.0-ref",
-  "seedance-2.0-ref-fast",
-  "kling3-standard",
-  "kling3-pro",
-  "kling3-4k",
-  "kling-o3-4k",
-  "happy-horse-t2v",
-  "happy-horse-i2v"
-]New value: +[
+  "seedance-2.0-t2v",
+  "seedance-2.0-t2v-fast",
+  "seedance-2.0-i2v",
+  "seedance-2.0-i2v-fast",
+  "seedance-2.0-ref",
+  "seedance-2.0-ref-fast",
+  "seedance-2.0-t2v-mini",
+  "seedance-2.0-i2v-mini",
+  "seedance-2.0-ref-mini",
+  "kling3-standard",
+  "kling3-pro",
+  "kling3-4k",
+  "kling-o3-4k",
+  "happy-horse-t2v",
+  "happy-horse-i2v"
+]
    • Changedgenerate_video_to_storyboard1 field changed
      • changedInput schema / properties / model / enum
        Previous value: -[
-  "seedance-2.0-t2v",
-  "seedance-2.0-t2v-fast",
-  "seedance-2.0-i2v",
-  "seedance-2.0-i2v-fast",
-  "seedance-2.0-ref",
-  "seedance-2.0-ref-fast",
-  "kling3-standard",
-  "kling3-pro",
-  "kling3-4k",
-  "kling-o3-4k",
-  "happy-horse-t2v",
-  "happy-horse-i2v"
-]New value: +[
+  "seedance-2.0-t2v",
+  "seedance-2.0-t2v-fast",
+  "seedance-2.0-i2v",
+  "seedance-2.0-i2v-fast",
+  "seedance-2.0-ref",
+  "seedance-2.0-ref-fast",
+  "seedance-2.0-t2v-mini",
+  "seedance-2.0-i2v-mini",
+  "seedance-2.0-ref-mini",
+  "kling3-standard",
+  "kling3-pro",
+  "kling3-4k",
+  "kling-o3-4k",
+  "happy-horse-t2v",
+  "happy-horse-i2v"
+]
  11. 2 tool updates
    • Addedgenerate_image_to_flow
    • Addedgenerate_video_to_flow
  12. 1 tool update
    • Changedgenerate_video_to_storyboard4 fields changed
      • changedInput schema / properties / duration / description
        Previous value: -"Video duration in seconds. Per-model bounds: seedance i2v/ref 4-15, seedance t2v 5-15, kling3-standard/pro 5-10, kling3-4k/o3-4k 3-15. Defaults to 5."New value: +"Video duration in seconds. Per-model bounds: seedance i2v/ref 4-15, seedance t2v 5-15, kling3-standard/pro 5-10, kling3-4k/o3-4k 3-15, happy-horse t2v/i2v 3-15. Defaults to 5."
      • changedInput schema / properties / model / description
        Previous value: -"Model. Defaults to 'seedance-2.0-t2v'. Use the -i2v variant or any kling3 variant for image-to-video, or seedance-2.0-ref / -ref-fast for reference-to-video."New value: +"Model. Defaults to 'seedance-2.0-t2v'. Use the -i2v variant, any kling3 variant, or happy-horse-i2v for image-to-video; happy-horse-t2v for Happy Horse text-to-video; or seedance-2.0-ref / -ref-fast for reference-to-video."
      • changedInput schema / properties / model / enum
        Previous value: -[
-  "seedance-2.0-t2v",
-  "seedance-2.0-t2v-fast",
-  "seedance-2.0-i2v",
-  "seedance-2.0-i2v-fast",
-  "seedance-2.0-ref",
-  "seedance-2.0-ref-fast",
-  "kling3-standard",
-  "kling3-pro",
-  "kling3-4k",
-  "kling-o3-4k"
-]New value: +[
+  "seedance-2.0-t2v",
+  "seedance-2.0-t2v-fast",
+  "seedance-2.0-i2v",
+  "seedance-2.0-i2v-fast",
+  "seedance-2.0-ref",
+  "seedance-2.0-ref-fast",
+  "kling3-standard",
+  "kling3-pro",
+  "kling3-4k",
+  "kling-o3-4k",
+  "happy-horse-t2v",
+  "happy-horse-i2v"
+]
      • changedInput schema / properties / resolution / description
        Previous value: -"Video resolution. Only meaningful for seedance (480p/720p/1080p; 1080p not allowed with seedance fast/ref-fast). Kling models lock resolution by variant."New value: +"Video resolution. Meaningful for seedance (480p/720p/1080p; 1080p not allowed with seedance fast/ref-fast) and happy-horse (720p/1080p, default 720p). Kling models lock resolution by variant."
  13. 1 tool update
    • Changedgenerate_video4 fields changed
      • changedInput schema / properties / duration / description
        Previous value: -"Video duration in seconds. Per-model bounds: seedance i2v/ref 4-15, seedance t2v 5-15, kling3-standard/pro 5-10, kling3-4k/o3-4k 3-15. Defaults to 5."New value: +"Video duration in seconds. Per-model bounds: seedance i2v/ref 4-15, seedance t2v 5-15, kling3-standard/pro 5-10, kling3-4k/o3-4k 3-15, happy-horse t2v/i2v 3-15. Defaults to 5."
      • changedInput schema / properties / model / description
        Previous value: -"Model. Defaults to 'seedance-2.0-t2v'. Use the -i2v variant or any kling3 variant for image-to-video, or seedance-2.0-ref / -ref-fast for reference-to-video."New value: +"Model. Defaults to 'seedance-2.0-t2v'. Use the -i2v variant, any kling3 variant, or happy-horse-i2v for image-to-video; happy-horse-t2v for Happy Horse text-to-video; or seedance-2.0-ref / -ref-fast for reference-to-video."
      • changedInput schema / properties / model / enum
        Previous value: -[
-  "seedance-2.0-t2v",
-  "seedance-2.0-t2v-fast",
-  "seedance-2.0-i2v",
-  "seedance-2.0-i2v-fast",
-  "seedance-2.0-ref",
-  "seedance-2.0-ref-fast",
-  "kling3-standard",
-  "kling3-pro",
-  "kling3-4k",
-  "kling-o3-4k",
-  "grok-imagine-video-v1-5"
-]New value: +[
+  "seedance-2.0-t2v",
+  "seedance-2.0-t2v-fast",
+  "seedance-2.0-i2v",
+  "seedance-2.0-i2v-fast",
+  "seedance-2.0-ref",
+  "seedance-2.0-ref-fast",
+  "kling3-standard",
+  "kling3-pro",
+  "kling3-4k",
+  "kling-o3-4k",
+  "grok-imagine-video-v1-5",
+  "happy-horse-t2v",
+  "happy-horse-i2v"
+]
      • changedInput schema / properties / resolution / description
        Previous value: -"Video resolution. Only meaningful for seedance (480p/720p/1080p; 1080p not allowed with seedance fast/ref-fast). Kling models lock resolution by variant."New value: +"Video resolution. Meaningful for seedance (480p/720p/1080p; 1080p not allowed with seedance fast/ref-fast) and happy-horse (720p/1080p, default 720p). Kling models lock resolution by variant."
  14. 2 tool updates
    • Changedgenerate_video12 fields changed
      • changedInput schema / properties / duration / description
        Previous value: -"Video duration in seconds. Per-model bounds: seedance i2v 4-15, seedance t2v 5-15, kling3-standard/pro 5-10, kling3-4k/o3-4k 3-15. Defaults to 5."New value: +"Video duration in seconds. Per-model bounds: seedance i2v/ref 4-15, seedance t2v 5-15, kling3-standard/pro 5-10, kling3-4k/o3-4k 3-15. Defaults to 5."
      • changedInput schema / properties / fast_mode / description
        Previous value: -"Legacy alias. true picks seedance-2.0-t2v-fast or seedance-2.0-i2v-fast when no explicit model was given. Prefer setting model directly."New value: +"Legacy alias. true picks the -fast variant (t2v/i2v/ref) when no explicit model was given. Prefer setting model directly."
      • changedInput schema / properties / file_id / description
        Previous value: -"file_id from prepare_image_upload — preferred for chat attachments. Required for seedance-2.0-i2v / -i2v-fast. Optional for kling3-* (presence triggers image-to-video mode)."New value: +"file_id from prepare_image_upload — preferred for chat attachments. Required for seedance-2.0-i2v / -i2v-fast. Optional for kling3-* (presence triggers image-to-video mode). Not used by reference-to-video — use reference_file_ids."
      • changedInput schema / properties / image_url / description
        Previous value: -"HTTPS URL of the source image. Use only if you already have a public URL; otherwise call prepare_image_upload and pass file_id."New value: +"HTTPS URL of the source image. Use only if you already have a public URL; otherwise call prepare_image_upload and pass file_id. Not used by reference-to-video — use reference_image_urls."
      • changedInput schema / properties / model / description
        Previous value: -"Model. Defaults to 'seedance-2.0-t2v'. Use the -i2v variant or any kling3 variant for image-to-video."New value: +"Model. Defaults to 'seedance-2.0-t2v'. Use the -i2v variant or any kling3 variant for image-to-video, or seedance-2.0-ref / -ref-fast for reference-to-video."
      • changedInput schema / properties / model / enum
        Previous value: -[
-  "seedance-2.0-t2v",
-  "seedance-2.0-t2v-fast",
-  "seedance-2.0-i2v",
-  "seedance-2.0-i2v-fast",
-  "kling3-standard",
-  "kling3-pro",
-  "kling3-4k",
-  "kling-o3-4k",
-  "grok-imagine-video-v1-5"
-]New value: +[
+  "seedance-2.0-t2v",
+  "seedance-2.0-t2v-fast",
+  "seedance-2.0-i2v",
+  "seedance-2.0-i2v-fast",
+  "seedance-2.0-ref",
+  "seedance-2.0-ref-fast",
+  "kling3-standard",
+  "kling3-pro",
+  "kling3-4k",
+  "kling-o3-4k",
+  "grok-imagine-video-v1-5"
+]
      • changedInput schema / properties / prompt / description
        Previous value: -"Text description of the video. For image-to-video, describe the motion/action you want applied to the source image."New value: +"Text description of the video. For image-to-video, describe the motion/action you want applied to the source image. For reference-to-video, describe the scene; the reference images lock the character/style."
      • addedInput schema / properties / reference_audio_urls
        Added value: +{
+  "description": "Reference-to-video only. HTTPS URLs of reference audio (max 3). Optional. Combined references cap at 12 total assets.",
+  "items": {
+    "type": "string"
+  },
+  "type": "array"
+}
      • addedInput schema / properties / reference_file_ids
        Added value: +{
+  "description": "Reference-to-video only. file_ids from prepare_image_upload, resolved to image references (counted alongside reference_image_urls toward the max of 9). Use for chat attachments.",
+  "items": {
+    "type": "string"
+  },
+  "type": "array"
+}
      • addedInput schema / properties / reference_image_urls
        Added value: +{
+  "description": "Reference-to-video only (seedance-2.0-ref / -ref-fast). HTTPS URLs of reference images that lock character/style. Storyboard Studio assets already have permanent URLs you can pass directly. At least one image reference is required; max 9.",
+  "items": {
+    "type": "string"
+  },
+  "type": "array"
+}
      • addedInput schema / properties / reference_video_urls
        Added value: +{
+  "description": "Reference-to-video only. HTTPS URLs of reference videos (max 3). Optional supplement to image references.",
+  "items": {
+    "type": "string"
+  },
+  "type": "array"
+}
      • changedInput schema / properties / resolution / description
        Previous value: -"Video resolution. Only meaningful for seedance (480p/720p/1080p; 1080p not allowed with seedance fast). Kling models lock resolution by variant."New value: +"Video resolution. Only meaningful for seedance (480p/720p/1080p; 1080p not allowed with seedance fast/ref-fast). Kling models lock resolution by variant."
    • Changedgenerate_video_to_storyboard11 fields changed
      • changedInput schema / properties / duration / description
        Previous value: -"Video duration in seconds. Per-model bounds: seedance i2v 4-15, seedance t2v 5-15, kling3-standard/pro 5-10, kling3-4k/o3-4k 3-15. Defaults to 5."New value: +"Video duration in seconds. Per-model bounds: seedance i2v/ref 4-15, seedance t2v 5-15, kling3-standard/pro 5-10, kling3-4k/o3-4k 3-15. Defaults to 5."
      • changedInput schema / properties / file_id / description
        Previous value: -"file_id from prepare_image_upload — preferred for chat attachments. Required for seedance-2.0-i2v / -i2v-fast. Optional for kling3-* (presence triggers image-to-video mode)."New value: +"file_id from prepare_image_upload — preferred for chat attachments. Required for seedance-2.0-i2v / -i2v-fast. Optional for kling3-* (presence triggers image-to-video mode). Not used by reference-to-video — use reference_file_ids."
      • changedInput schema / properties / image_url / description
        Previous value: -"HTTPS URL of the source image. Use only if you already have a public URL; otherwise call prepare_image_upload and pass file_id."New value: +"HTTPS URL of the source image. Use only if you already have a public URL; otherwise call prepare_image_upload and pass file_id. Not used by reference-to-video — use reference_image_urls."
      • changedInput schema / properties / model / description
        Previous value: -"Model. Defaults to 'seedance-2.0-t2v'. Use the -i2v variant or any kling3 variant for image-to-video."New value: +"Model. Defaults to 'seedance-2.0-t2v'. Use the -i2v variant or any kling3 variant for image-to-video, or seedance-2.0-ref / -ref-fast for reference-to-video."
      • changedInput schema / properties / model / enum
        Previous value: -[
-  "seedance-2.0-t2v",
-  "seedance-2.0-t2v-fast",
-  "seedance-2.0-i2v",
-  "seedance-2.0-i2v-fast",
-  "kling3-standard",
-  "kling3-pro",
-  "kling3-4k",
-  "kling-o3-4k"
-]New value: +[
+  "seedance-2.0-t2v",
+  "seedance-2.0-t2v-fast",
+  "seedance-2.0-i2v",
+  "seedance-2.0-i2v-fast",
+  "seedance-2.0-ref",
+  "seedance-2.0-ref-fast",
+  "kling3-standard",
+  "kling3-pro",
+  "kling3-4k",
+  "kling-o3-4k"
+]
      • changedInput schema / properties / prompt / description
        Previous value: -"Text description of the video. For image-to-video, describe the motion/action you want applied to the source image."New value: +"Text description of the video. For image-to-video, describe the motion/action you want applied to the source image. For reference-to-video, describe the scene; the reference images lock the character/style."
      • addedInput schema / properties / reference_audio_urls
        Added value: +{
+  "description": "Reference-to-video only. HTTPS URLs of reference audio (max 3). Optional. Combined references cap at 12 total assets.",
+  "items": {
+    "type": "string"
+  },
+  "type": "array"
+}
      • addedInput schema / properties / reference_file_ids
        Added value: +{
+  "description": "Reference-to-video only. file_ids from prepare_image_upload, resolved to image references (counted alongside reference_image_urls toward the max of 9). Use for chat attachments.",
+  "items": {
+    "type": "string"
+  },
+  "type": "array"
+}
      • addedInput schema / properties / reference_image_urls
        Added value: +{
+  "description": "Reference-to-video only (seedance-2.0-ref / -ref-fast). HTTPS URLs of reference images that lock character/style. Storyboard character/location assets already have permanent URLs you can pass directly. At least one image reference is required; max 9.",
+  "items": {
+    "type": "string"
+  },
+  "type": "array"
+}
      • addedInput schema / properties / reference_video_urls
        Added value: +{
+  "description": "Reference-to-video only. HTTPS URLs of reference videos (max 3). Optional supplement to image references.",
+  "items": {
+    "type": "string"
+  },
+  "type": "array"
+}
      • changedInput schema / properties / resolution / description
        Previous value: -"Video resolution. Only meaningful for seedance (480p/720p/1080p; 1080p not allowed with seedance fast). Kling models lock resolution by variant."New value: +"Video resolution. Only meaningful for seedance (480p/720p/1080p; 1080p not allowed with seedance fast/ref-fast). Kling models lock resolution by variant."
  15. 2 tool updates
    • Changedgenerate_image_to_storyboard3 fields changed
      • addedInput schema / properties / beat_number
        Added value: +{
+  "description": "For role 'beat': the beat's position in story order (1..N). Used to keep frames and clips ordered in the Library.",
+  "minimum": 1,
+  "type": "integer"
+}
      • addedInput schema / properties / label
        Added value: +{
+  "description": "Short human label shown in the Studio Library, e.g. 'Pip - hero character' or 'Beat 3 - Into the Backrooms'.",
+  "type": "string"
+}
      • addedInput schema / properties / role
        Added value: +{
+  "description": "Studio Library grouping for storyboard work. 'cast' = a character reference, 'location' = a place/world reference, 'beat' = a single storyboard frame (still) for a beat. Omit for a plain generated image.",
+  "enum": [
+    "cast",
+    "location",
+    "beat"
+  ],
+  "type": "string"
+}
    • Changedgenerate_video_to_storyboard2 fields changed
      • addedInput schema / properties / beat_number
        Added value: +{
+  "description": "The beat's position in story order (1..N). Keeps generated clips ordered in the Library to match the storyboard.",
+  "minimum": 1,
+  "type": "integer"
+}
      • addedInput schema / properties / label
        Added value: +{
+  "description": "Short human label shown in the Studio Library, e.g. 'Beat 3 - Into the Backrooms'.",
+  "type": "string"
+}
  16. 2 tool updates
    • Addedextract_brand
    • Addedresearch_business

Related MCP Servers

  • Citesurf MCP

    citesurf
    A
    license
    A
    quality
    C
    maintenance
    Enables brand visibility monitoring across major AI platforms like ChatGPT, Claude, Gemini, and Perplexity. It allows users to track visibility scores, analyze competitor data, and receive actionable insights to improve AI-generated brand recommendations.
    16
    7 npm
    1
    MIT

  • CompetitorPulse

    vk-ai
    A
    license
    Not graded
    quality
    B
    maintenance
    Enables tracking competitor websites, changelogs, blog feeds, and pricing pages with meaningful diffs, classification, and Markdown digests via MCP tools for listing, adding, removing competitors, running checks, and retrieving digests or changes.
    MIT
  • industrylens-mcp by IndustryLensOp

    industrylens-mcpofficial

    IndustryLensOp
    A
    license
    Not graded
    quality
    B
    maintenance
    Browse IndustryLens's published competitive-intelligence reports and head-to-head competitor comparisons from any AI agent — real, source-backed data.
    MIT
Try in Browser
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources