NeoHotaru/dsh-read-animation

dsh-read-animation lets the agent actually see an animated GIF: instead of one first frame from read_image, it decodes every frame and hands the model a labelled sequence of real frames sampled across the animation's timeline.

dsh-read-animation lets a DeepSeek Harness agent actually see an animated GIF. Reading a GIF with read_image normally gives the model one picture - the first frame - with no signal that the other frames existed; this plugin decodes every frame and hands the model a labelled sequence of real frames sampled across the animation's timeline, so the model can reason about the motion instead of a single still.

Vision & Multimodal ★ 0 updated 2026-10-15
View on GitHub ↗

Install

dsh plugin --profile web add dsh-read-animation

Copied verbatim from the README's install block; the README also documents a tagged-commit form (github:NeoHotaru/dsh-read-animation#v0.1.0) and a link:$PWD local-development form.

Compatibility

Ships as a DeepSeek Harness bundle; installing it registers the bundle, and the README notes the sharp native dependency is handled by the install. English README.

Details

Alternatives

54xkeee/dsh-vision · 54xkeee/dsh-youreyes · Cheng-cheng9669/dsh-deepseek-vision

More plugins in Vision & Multimodal

Browse more in Vision & Multimodal

Guides for Vision & Multimodal plugins