The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to search for application binaries using system search commands.- Evidence: SKILL.md recommends using command -v, where, and mdfind to locate cadcraft and cadcraft-cli on the host system.- [EXTERNAL_DOWNLOADS]: The skill references protocol documentation hosted on a public repository.- Evidence: SKILL.md provides a link to https://raw.githubusercontent.com/storytold/cadcraft/refs/heads/main/docs/control-protocol.md to define the control interface.- [INDIRECT_PROMPT_INJECTION]: The skill processes external CAD files, creating a vulnerability surface for indirect prompt injection.- Ingestion points: The app.open method in SKILL.md and references/control-protocol.md allows the loading of arbitrary external CAD files.- Boundary markers: There are no instructions for the agent to distinguish between drawing data and potential instructions embedded within the file.- Capability inventory: The skill can execute CAD commands (engine.execute), enter command-line text (cmdline.input), and perform file operations or rendering (app.save, ui.render, ui.screenshot), as detailed in references/control-protocol.md.- Sanitization: The documentation does not specify any sanitization or validation of the ingested CAD content.

cadcraft