Skip to main content
Glama

Edit by transcript

edit_transcript

Edit by text: strike or restore word ranges (inclusive indices), clean_up (strike every filler and shorten pauses over 0.6s to 0.3s), set the layout from a word onwards (screen, camera, screen+bubble, side-by-side, stack; for a phone take camera+phone, with side left or right, and phone), pick a scene's take, burn in captions, end card on or off. Returns the new finished length. Zoom the screen recording: add_zoom from_word..to_word on a take (level 1.1 to 3, default 1.5; center 0-1, default the middle), update_zoom, remove_zoom. get_project lists zooms with their level after the sharpness limit and any that are hidden or under the camera bubble. add_zoom returns added_zoom_ids.

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>",
      "actions": "<actions>"
    },
    "name": "edit_transcript"
  }
}

Input Schema

NameRequiredDescriptionDefault
actionsYes
project_idYesFrom create_project

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observed

TDQS

B3.1/5.0
Behavior3/5

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

Annotations already declare readOnly=false, destructive=false, idempotent=false, so the safety baseline is covered. The description usefully adds return behavior ("Returns the new finished length", "add_zoom returns added_zoom_ids") and discloses what clean_up actually does (strikes fillers, shortens pauses over 0.6s to 0.3s), which is real behavioral content. It still omits whether a batch is atomic, what happens on partial failure, and whether 4 ops are similarly destructive.

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

Conciseness2/5

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

Two extremely long sentences bundle nine operations with no line breaks, bullets, or front-loading; the most common op (strike/restore) is not visually separated from niche zoom details. The zoom block at the end reads as an afterthought and the parenthetical qualifiers interrupt the flow repeatedly, making the agent parse rather than scan.

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 a single tool fronting nine variant actions with no output schema, the description is unusually complete — it covers every op's payload at a semantic level and even reports return shapes. Remaining gaps are the missing batch-size limit (1–200), atomicity semantics, and the relationship between scene_id and take_id, which would matter for real invocation correctness.

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 coverage is 50% — only project_id carries a description — so the description has to carry the inner action fields, and it does meaningfully: inclusive indices for from/to, from_word..to_word typing, level range 1.1–3 with default 1.5, center 0–1 defaulting to the middle, and enum-style guidance for phone layouts. It does not explain take_id vs scene_id, nor the schema's 1–200 actions cap, so it adds value without fully compensating.

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?

The description opens with a specific verb+resource ("Edit by text") and then enumerates the concrete operations: strike/restore word ranges, clean_up, layout, select_take, captions, end_card, and the three zoom ops. It distinguishes the tool from siblings like get_transcript or add_graphic by being clearly the mutation path for transcript-level edits. It loses a point only because the operations are crammed into two run-on sentences rather than listed as a scannable set.

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

Usage Guidelines2/5

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

There is one implicit workflow hint — get_project is named as the way to discover existing zooms and their levels — but no explicit statement of when to call edit_transcript versus get_transcript or add_take. No preconditions, no exclusions, no guidance on when a batch of actions is preferable to separate calls.

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