PichurChill

PichurChill/j-can-see

Skill Claude CodeCodex

Vision methodology and file-output conventions for the j-can-see MCP server. TRIGGER GATE — on models WITH native vision input, invoke this skill only when the user explicitly asks for j-can-see, or when native vision has actually failed on the task (cannot open/read an image, unusable output). Otherwise read images…

PichurChill/j-can-see

MCP server Claude CodeCodexCursor +2

MCP server that gives text-only AI coding agents a vision toolkit: describe/OCR images, locate elements with pixel coordinates, diff images, pick exact colors, trace SVGs — for clients whose primary model has no multimodal input. Runs locally from the j-can-see npm package.