Skip to main content
Glama

Dock

by ai.trydock
OverviewDiscussionsAdmin

watch_video

Look at a video in a video surface: returns frames as images, each labelled with its time, plus what is said in that span (transcribed and cached on first use). With no position you get an overview spread across the whole video. When someone points at a moment ("I don't like 1:34"), pass at: "1:34" to see the seconds around it; start/end show a range; times shows exact moments. Inside a span, frames sets how many, and fps sets a rate: the video's own frame rate over a short window shows every single frame. Up to 32 frames per call. Look at the overview first, then zoom in. Viewer role required.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
atNoA moment to look around. As a person writes it ("1:34", "1:34.5", "01:02:03") or as seconds (94).
endNoEnd of a range. Defaults to the end of the video. As a person writes it ("1:34", "1:34.5", "01:02:03") or as seconds (94).
fpsNoFrames per second across the span instead of a count. Use the video's fps (returned in `video.fps`) to see every frame.
sizeNoFrame size: small 512px, medium 768px (default), large 1280px on the long side.
slugYesThe workspace slug. Accepts bare or org-prefixed form.
startNoStart of a range. As a person writes it ("1:34", "1:34.5", "01:02:03") or as seconds (94).
timesNoExact moments to show, e.g. ["0:12", "1:34.2"].
framesNoHow many frames across the span. Default 12 for an overview or range, 9 around `at`. Max 32.
windowNoWith `at`: seconds to show on each side. Default 2.
video_idNoThe video's id, from list_videos. Optional when the surface holds a single video.
comment_idNoOpen a video comment: shows the moment it was left at (1 s each side) with the box the person drew outlined in red, plus what they wrote. Overrides surface_slug, video_id and the position args. Find them with list_comments target_type video_frame.
surface_slugNoOptional video surface slug. Omit for the first video surface.
include_transcriptNoInclude what is said in the span. Default true.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Added

TDQS

A4.4/5.0
Behavior4/5

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

With no annotations, the description carries the full burden and does well: it discloses that transcripts are cached on first use, that the call is capped at 32 frames, that a viewer role is required, and that `comment_id` overrides surface_slug/video_id/position args and outlines the drawn box in red. It stops short of describing error conditions or performance/staleness trade-offs.

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 core behavior in sentence one, then modes, then limits and prerequisites. Dense with parentheticals but each sentence is functional; appropriate length given 13 parameters and no annotations to lean on.

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?

No output schema exists, so the description must convey return shape, which it does (frames as time-labelled images plus spoken span), along with limits, prerequisites, override semantics, and default behavior. An agent can call this tool correctly from the description alone.

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 100% so baseline is 3, but the description adds real meaning: it explains the interaction of `at` vs `start`/`end` vs `times`, that `frames` counts while `fps` sets a rate, that the video's own fps reveals every frame, and that `comment_id` overrides other position args. This improves on the schema without contradicting it.

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?

States a specific verb+resource ('Look at a video in a video surface') and immediately defines the output ('returns frames as images, each labelled with its time, plus what is said in that span'). This distinguishes it clearly from siblings like get_video_transcript (transcript only) and edit_video/update_video (mutations).

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?

Gives explicit mode selection: no position = whole-video overview, `at` for around a moment, `start`/`end` for a range, `times` for exact moments, plus the workflow 'Look at the overview first, then zoom in' and the 'Viewer role required' prerequisite. It does not, however, explicitly name alternative siblings such as get_video_transcript for the text-only case.

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

All Dock toolsInspect
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