Skip to main content
Glama
Hermoso

Hermoso

Official

Insert a new scene into a video

insert_scene

Insert a newly generated 5-13 second scene into an existing clip at any point, then resume the original footage without re-rendering the rest. Use it for product close-ups, reactions, or reveals mid-ad.

Instructions

INSERT a newly generated scene into a clip you already have: the video plays up to startTime, the new scene (5 to 13 seconds) plays, then the original picks up again at resumeTime. Use for a product close-up, a reaction, a reveal or a missing beat in the middle of an ad without re-rendering the rest. resumeTime = startTime keeps every second of the original; a later resumeTime also cuts the footage in between. Describe the scene in prompt and/or attach up to 4 reference pictures (refImages) or 1 reference clip (refVideos). Clips up to 60 seconds; both times between 1.625 and 60 seconds. Paid render (billed per second of NEW scene); returns the whole video's URL.

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": {
      "video": "<video>",
      "startTime": "<startTime>",
      "resumeTime": "<resumeTime>"
    },
    "name": "insert_scene"
  }
}

Input Schema

NameRequiredDescriptionDefault
videoYesthe source video URL (a render, job result or list_library item)
promptNowhat happens in the new scene: the action, the subject, the camera
secondsNolength of the new scene, 5-13 seconds (default 5)
get_costNotrue: return the price only; nothing runs or bills
refImagesNoup to 4 picture URLs of what should appear (the product, a person, a place)
refVideosNoup to 1 clip URL whose look or motion the scene follows
startTimeYesseconds into the clip where the new scene starts (1.625-60)
resolutionNodefault 720p (rendered at 768p); '480p' is cheaper
resumeTimeYesseconds into the ORIGINAL clip where it picks up again after the new scene (2.625-60, at or after startTime)
idempotency_keyNochosen once, resent unchanged on retries: the same key never buys twice
expected_creditsNothe get_cost price; refused unbilled if the live price differs

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed3 schema fields changedv0.1.532
    • addedInput schema / properties / expected_credits
      Added value: +{
      +  "description": "the get_cost price; refused unbilled if the live price differs",
      +  "type": "number"
      +}
    • addedInput schema / properties / get_cost
      Added value: +{
      +  "description": "true: return the price only; nothing runs or bills",
      +  "type": "boolean"
      +}
    • addedInput schema / properties / idempotency_key
      Added value: +{
      +  "description": "chosen once, resent unchanged on retries: the same key never buys twice",
      +  "type": "string"
      +}
  2. Addedv0.1.454

TDQS

A4.2/5.0
Behavior3/5

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

Beyond the annotations it discloses the paid-render billing model (per second of new scene), the returned whole-video URL, the cost-only dry run, and the important behavioral note that a later resumeTime silently drops the in-between footage. That last point sits in mild tension with destructiveHint=false but is best read as output trimming rather than data destruction. The idempotentHint=false annotation plus the opt-in idempotency_key is coherent, not contradictory.

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?

Front-loaded with the action and splice sequence, then constraints and cost. It is a dense single paragraph with no wasted sentences, though the middle section mixing limits, references and billing could be broken into shorter units for faster scanning.

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

Completeness5/5

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

For an 11-parameter, non-read-only render tool with no output schema, it covers what an agent needs: duration and time bounds, reference limits, resolution tradeoff, cost preview, idempotency for retries, and the shape of the return value. Nothing material is missing.

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 coverage is already 100%, so the baseline is 3, but the description adds real meaning the schema does not: the relational rule "resumeTime = startTime keeps every second; a later resumeTime also cuts the footage in between" and the 5–13 second scene length. This is genuine semantic value over the field-level text.

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?

"INSERT a newly generated scene into a clip you already have" states a specific verb and resource, and the follow-up sentence explains the exact splice mechanics (play to startTime, new scene, resume at resumeTime). An agent can distinguish this from whole-video generation or video-editing siblings 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 names concrete use cases (product close-up, reaction, reveal, missing beat mid-ad) and the key framing of inserting "without re-rendering the rest," which is what separates it from edit/generate alternatives. It stops short of naming a sibling tool explicitly or stating when not to use it.

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

Deploy Server

Other Tools