screenshotlisted

<purpose> Automatically capture a screenshot because the conversation context suggests the user needs one. This skill is invoked proactively -- do not ask the user questions, just capture and show the result. Invoke this skill when the user says things like: - "let me see what that looks like" - "capture the current state" - "screenshot this" - "show me the screen" - "take a screenshot of [app]" </purpose> <steps> <step id="load-config" number="1"> <description>Load Config</description> <load-config> <action>Read `.claude/screenshotr.local.md` for settings.</action> <if condition="no-config"> <action>Capture with sensible defaults: png format, `./screenshots` directory, no shadow, silent.</action> </if> </load-config> </step> <step id="detect-target" number="2"> <description>Detect Target</description> <action>Infer the capture target from conversation context.</action> <if condition="specific-app-discussed"> <action>Capture that app's window using `bash "${CLAUDE_PLUGIN_ROOT}/scripts/get-window-id.sh" "<AppName>"`.</action> </if> <if condition="no-specific-app"> <action>Fullscreen capture.</action> </if> <if condition="user-mentioned-url"> <action>This is better handled by `/capture-ss url "..."` -- suggest that instead and stop.</action> <exit /> </if> </step> <step id="generate-filename" number="3"> <description>Generate Filename</description> <action>Cr