modlens
A vision plugin that gives text-only coding agents structured image evidence.
Free
Development & Engineering
★ 4848
#138
of 289 skill repos
4.8k
GitHub stars
liustack
author
// about
DeepSeek's flagship chat models and GLM-5.3 can't read images. ModLens is a plug-in vision engine: paste an image and the agent gets structured JSON evidence covering OCR, layout and semantics. It targets DeepSeek Harness first and any text-only coding agent second.
// how to install
Run npx -y skills add liustack/modlens --skill modlens --global.
// works with
Claude CodeCodex
// similar skills in Development & Engineering
rea
Reverse engineering with agents, from app behavior down to native binaries.
morluto · Development & Engineering · ★ 441806
superpowers
A Claude Code plugin bundling skills for TDD, debugging, and planning workflows.
obra · Development & Engineering · ★ 312082
skills
Matt Pocock's personal collection of Agent Skills for everyday engineering work.
mattpocock · Development & Engineering · ★ 311987
ECC
A harness setup bundling skills, memory, and security rules for coding agents.
affaan-m · Development & Engineering · ★ 296776
ponytail
A skill that pushes an agent to write less code and reuse what already exists.
DietrichGebert · Development & Engineering · ★ 184054
agent-skills
25 Claude skills covering a full engineering lifecycle from requirements to shipping.
addyosmani · Development & Engineering · ★ 113136