Skip to main content
Glama

Contact Sheet

contact_sheet
Destructive

Tile timestamped video frames into a single image to review an entire recording at a glance and identify key moments.

Instructions

Tile timestamped frames of a video into one image to see a whole recording at a glance (the cheapest first look).

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": {
      "input": "<input>"
    },
    "name": "contact_sheet"
  }
}

Input Schema

NameRequiredDescriptionDefault
modeNoeven (default) or scenes (strongest changes; static stretches take no tiles).
countNoFrames to tile (default 12).
inputYesPath to the source video.
outputNoOptional output .png path. Defaults under SCREENCAST_HOME/frames.
posterNoWrite one representative poster frame instead of a sheet.
columnsNoTiles per row (default: roughly square).
overwriteNoAllow replacing an existing file at the output path (default false).
tileWidthNoTile width in px (default 320).
returnImageNoReturn the sheet inline (default true).
maxImageWidthNoMax inline image width (default 1280).
sceneThresholdNoMode scenes: change needed, 0-1 (default 0.1).

Output Schema

NameRequiredDescriptionDefault
outputPathYes
inlineImagesYes

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed8 schema fields changedv0.47.0
    • changedInput schema / properties / count / description
      Previous value: -"How many frames to tile (default 12, max 48)."New value: +"Frames to tile (default 12)."
    • changedInput schema / properties / maxImageWidth / description
      Previous value: -"Downscale the inline image to at most this width (default 1280)."New value: +"Max inline image width (default 1280)."
    • changedInput schema / properties / mode / description
      Previous value: -"'even' spreads frames evenly across the video (default); 'scenes' uses the opening frame plus the strongest scene changes, so static stretches take no tiles."New value: +"even (default) or scenes (strongest changes; static stretches take no tiles)."
    • changedInput schema / properties / poster / description
      Previous value: -"Instead of a sheet, write one representative poster frame (ffmpeg's thumbnail filter over 30 candidates spread across the video), e.g. for a video thumbnail."New value: +"Write one representative poster frame instead of a sheet."
    • changedInput schema / properties / returnImage / description
      Previous value: -"Also return the sheet inline as an image the model can view directly (default true)."New value: +"Return the sheet inline (default true)."
    • changedInput schema / properties / sceneThreshold / description
      Previous value: -"Mode 'scenes': how much a frame must change to count (default 0.1)."New value: +"Mode scenes: change needed, 0-1 (default 0.1)."
    • changedInput schema / properties / tileWidth / description
      Previous value: -"Width of each tile in px (default 320)."New value: +"Tile width in px (default 320)."
    • changedOutput schema / (root)
      Previous value: -nullNew value: +{
      +  "$schema": "http://json-schema.org/draft-07/schema#",
      +  "additionalProperties": {},
      +  "properties": {
      +    "inlineImages": {
      +      "maximum": 9007199254740991,
      +      "minimum": -9007199254740991,
      +      "type": "integer"
      +    },
      +    "outputPath": {
      +      "type": "string"
      +    }
      +  },
      +  "required": [
      +    "outputPath",
      +    "inlineImages"
      +  ],
      +  "type": "object"
      +}
  2. First observedv0.32.0

TDQS

B3.3/5.0
Behavior2/5

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

Annotations declare destructiveHint=true and readOnlyHint=false, yet the description never explains why: it writes a PNG to disk, has an overwrite guard defaulting to false, and can emit a poster frame instead of a sheet. Those are precisely the behaviors an agent needs before calling a destructive tool, and the description adds none of them.

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

Conciseness5/5

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

One sentence, front-loaded with the action and result, with the parenthetical qualifier earning its place by signaling cost. There is no filler.

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

Completeness3/5

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

An output schema and full schema descriptions cover return values and parameters, so the description needn't restate them. But for an 11-parameter tool that writes files destructively, the absence of any note on disk output, the overwrite guard, or the even-vs-scenes tradeoff leaves a meaningful gap.

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

Parameters3/5

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

Schema description coverage is 100%, so every parameter including mode, count, poster, and overwrite is already documented in the schema. The description adds no syntax, defaults, or mode-selection guidance beyond what structured data provides, so baseline 3 applies.

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

Purpose4/5

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

States a specific verb and resource ('tile timestamped frames of a video into one image') plus the resulting artifact, so an agent knows exactly what is produced. It hints at its place in the workflow ('cheapest first look') but never names the nearest sibling, sample_frames, so differentiation is only implied.

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

Usage Guidelines3/5

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

'The cheapest first look' implies a first-pass triage use case, which is useful context. However, there is no explicit when-not guidance and no alternative named (sample_frames, review_recording, get_media_info), so the agent must infer the routing decision.

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