Skip to main content
Glama
ronak-create

FableCut

by ronak-create
OverviewSchemaSourceRelated ServersDiscussionsAdmin
JavaScript
Hybrid

fablecut_reframe

Reframe a sequence to another aspect ratio (9:16, 1:1, 4:5) while automatically following the subject with keyframes, so vertical and square edits stay framed without hand-tracking every shot.

Instructions

Reframe for another shape, following the subject: a 16:9 edit made 9:16 (Reels, Shorts, TikTok), 1:1 or 4:5. With aspect, a copy of the open sequence is made on that canvas (the project's shorter side kept, e.g. 1920×1080 → 1080×1920) and opened, so the original cut stays as it was; without aspect, the open sequence is reframed on its own canvas. Each full-frame shot (a video or still that isn't moved, scaled, cropped or animated by hand) is fitted to cover the canvas, and its position follows what matters in it: frames are sampled 5 times a second, the subject is found by motion against the shot's own movement, detail and colour, and a virtual camera operator holds still until the subject leaves a dead zone, then eases after it no faster than a pan speed. The result is ordinary x (or y) keyframes, thinned to a few, or one steady position — edit them like any keys. Shots without a clear subject are centred. Titles, picture-in-picture, clips placed by hand and locked clips are left alone (listed in the reply) — check them with fablecut_frame on the new canvas. clipIds limits it to some clips; follow:false only fits and centres; smooth 0…1 (default 0.5) — higher holds longer and pans slower. A version is saved first. Needs ffmpeg on PATH.

Input Schema

NameRequiredDescriptionDefault
nameNoWith aspect: the new sequence's name (default: the open one's name and the shape)
forceNoAlso reframe locked clips (only when the user asked)
aspectNoMake a copy of the open sequence on this canvas shape and reframe it (default: reframe the open sequence as it is)
followNoFollow the subject (default true); false only fits and centres
smoothNo0…1, default 0.5: higher holds longer and pans slower
clipIdsNoOnly these clips (default: every full-frame shot)

Schema Changelog

Changes observed during successful MCP inspections.

  1. Addedv1.17.0

TDQS

A4.5/5.0
Behavior5/5

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

With no annotations, the description carries the full burden and does: a version is saved first, the original cut stays intact when aspect is used, only full-frame shots are changed while titles/PiP/locked clips are left alone and listed in the reply, and ffmpeg on PATH is required. These are the exact side-effect and prerequisite details an agent needs on a mutation tool.

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

Conciseness3/5

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

Front-loaded correctly with the core action, but it is a single dense paragraph, and the algorithm internals (frames sampled 5 times a second, motion vs the shot's own movement, detail and colour) contribute little to tool selection. Structure would improve with the when-to-use and caveats split out.

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 a mutation tool with no annotations and no output schema, the description covers prerequisites (ffmpeg), side effects (version saved, copy created), what is skipped and how that is reported, and how to verify afterward via fablecut_frame. Nothing an agent needs before calling it 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 schema is the baseline. The description still adds useful semantics: 'follow:false only fits and centres' clarifies the mode, and 'smooth 0…1 — higher holds longer and pans slower' explains the dead-zone/pan-speed behavior rather than just the range.

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 first sentence names a specific verb (reframe) plus the resource and the goal (following the subject across 16:9→9:16, 1:1, 4:5), with concrete example shapes. It is distinguishable from every sibling because it is the only description that makes a new sequence on a different canvas.

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?

Explicitly states the branch between passing 'aspect' (creates a copy, original untouched) and omitting it (reframe the open sequence in place), and says 'force' should only be used when the user asked. It also points to the follow-up tool (fablecut_frame) for verifying clips it skipped, but gives no explicit when-not-to-use vs siblings.

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