Skip to main content
Glama

Change a still image

update_still

Changes a long video's still image: its words (at, until; until null for the default length), placement, inset, side, fit, movement, focus or fade, with the same checks as add_still; a new asset_id, image_url or image_file replaces the picture. upload: true gives a fresh upload_url only for a still that is still waiting for its picture (a link that ran out); for a still that has its picture, add a new still instead. Free. Returns the same as add_still.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "project_id": "<project_id>",
      "still_id": "<still_id>"
    },
    "name": "update_still"
  }
}

Input Schema

NameRequiredDescriptionDefault
atNoThe word it appears on, { take_id, word } from get_transcript
fitNoauto (default) fills its box when that keeps three quarters of the image, else shows it whole over a blurred copy; fill crops; whole shows all of it
fadeNoSeconds of fade in and out: 0 for a cut, or 0.1 to 0.5; default 0.3. A split doesn't fade
moveNoin (a slow push from 100% to 110%), out, or none; default in, none for an inset
sideNo
focusNo
insetNo
untilNoThe word it goes with; leave out (or null) for the default: the end of its sentence, at most 8 s
uploadNotrue: the person gives their own picture. The still is saved waiting, and you get upload_url for them (valid an hour). false is the same as leaving it out
asset_idNoAn image in the brand kit (get_brand lists its asset ids): the project's kit, or brand_id's
brand_idNoThe kit asset_id is in; default the project's
still_idYesFrom add_still or get_project
image_urlNoA ClipLaunch image link (a generate_assets stored_url, a compose_asset URL, a get_brand view_url), or a public https image the person gave you or owns. Fetched once, now
placementNofull (default) covers the frame; screen fills the screen's box under the camera bubble; inset is a rounded 16:9 box in a corner; split halves the frame, the camera on one side and the still on the other (the screen recording isn't shown)
at_secondsNoOr the moment in the finished video (what review_render shows), turned into the word playing then
image_fileNoChatGPT only: the image the person attached, as a list of one. Leave out in other apps
project_idYesFrom create_project
until_secondsNoOr the moment in the finished video it goes, turned into the last word started by then

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed2 schema fields changed
    • addedInput schema / properties / image_file
      Added value: +{
      +  "description": "ChatGPT only: the image the person attached, as a list of one. Leave out in other apps",
      +  "items": {
      +    "properties": {
      +      "download_url": {
      +        "maxLength": 4096,
      +        "type": "string"
      +      },
      +      "file_id": {
      +        "maxLength": 200,
      +        "type": "string"
      +      },
      +      "file_name": {
      +        "maxLength": 255,
      +        "type": "string"
      +      },
      +      "mime_type": {
      +        "maxLength": 100,
      +        "type": "string"
      +      }
      +    },
      +    "required": [
      +      "download_url",
      +      "file_id"
      +    ],
      +    "type": "object"
      +  },
      +  "maxItems": 1,
      +  "minItems": 1,
      +  "type": "array"
      +}
    • addedInput schema / properties / upload
      Added value: +{
      +  "description": "true: the person gives their own picture. The still is saved waiting, and you get upload_url for them (valid an hour). false is the same as leaving it out",
      +  "type": "boolean"
      +}
  2. Added

TDQS

A4.3/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Beyond the annotations (non-read-only, non-destructive, open-world) it discloses that the operation is free, that a new asset_id/image_url/image_file replaces the existing picture, that upload_url expires after an hour, and that remote images are fetched once at call time. The only unaddressed point is that unspecified fields appear to be left untouched, which for an 18-parameter updater is meaningful behavior. Replacing the picture is a mild tension with destructiveHint=false but is the normal semantics of an update, so this is not a contradiction.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Purpose and the main edit surface are front-loaded, and nearly every clause carries information (cost, upload semantics, picture replacement, return value). It is a very dense single block with heavy parenthetical nesting, and "with the same checks as add_still" plus "Returns the same as add_still" force the agent to look up another tool's definition.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For an 18-parameter updater with nested objects and no output schema, the description covers the tricky cases (word-pair timing, upload expiry, image source rules) and delegates return values via "Returns the same as add_still". The remaining gap is whether omitted parameters preserve existing values or reset to defaults, which an agent would want stated for a partial-update tool.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is already 83%, so the baseline is 3; the description adds real meaning on top by explaining the at/until word-pair semantics and the "until null for the default length" default, the at_seconds/until_seconds alternates, the accepted image sources for image_url, and the provenance of still_id (add_still or get_project). It stops short of explaining how at and at_seconds interact when both are supplied.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description opens with a specific verb and resource: "Changes a long video's still image", then enumerates exactly which attributes are editable (words, placement, inset, side, fit, movement, focus, fade) and which fields replace the picture. It explicitly distinguishes itself from the sibling add_still by scoping itself to an existing still, so an agent can route correctly without opening the schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It gives a concrete when-not: "upload: true gives a fresh upload_url only for a still that is still waiting for its picture... for a still that has its picture, add a new still instead", which names the alternative tool and the condition that selects it. What it lacks is broader framing of when to edit an existing still versus removing and re-adding one, and no guidance on update_still versus the update_graphic/update_scene siblings.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources