The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill dynamically fetches 'guides' from a remote server using the dpai guide get command, which are used as the primary logic for authoring tasks. \n
Ingestion points: External content is brought into the agent's context through dpai guide get, dpai context, and dpai design context as described in SKILL.md. \n
Boundary markers: The skill does not instruct the agent to use any delimiters or safety prompts to isolate these external instructions from its core system behavior. \n
Capability inventory: The agent can perform significant filesystem operations, including creating and patching files, as well as executing arbitrary CLI subcommands. \n
Sanitization: No input validation or sanitization of the remote guide content is specified, allowing remote data to potentially control agent actions. \n- [EXTERNAL_DOWNLOADS]: The skill setup process involves downloading and installing the desainpakeai-cli package from the global NPM registry. \n- [COMMAND_EXECUTION]: The skill relies on shell command execution to perform all its core functions, including authentication, project initialization, and file system mutations.

desainpakeai-prototype