The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read and act on output from the tdk doctor command. This creates a surface where malicious strings in the environment state or configuration could influence agent behavior.
Ingestion points: Diagnostic output from tdk doctor (SKILL.md).
Boundary markers: None specified to help the agent distinguish between data and instructions.
Capability inventory: Local execution of tdk CLI commands like doctor, networks, and up.
Sanitization: No sanitization or validation of the command output is requested before processing.
[COMMAND_EXECUTION]: The skill utilizes the tdk command-line utility for project setup and health checks. These commands are local development tools associated with the skill author.

tdk-doctor