The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute a local Python script ../../scripts/resolve_board_profile.py using user-provided input in the --query parameter. If the agent does not sanitize the input <board name> before passing it to the shell, it could lead to command injection.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external repository files, including ../../references/boards/index.json and Markdown profiles. These files represent an attack surface where malicious instructions could be embedded to influence the agent's behavior.
Ingestion points: Data is read from ../../references/boards/index.json and various Markdown files in the same directory.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are used when processing these files.
Capability inventory: The agent can execute a local Python script and hand off data to other skills like pin-assignment and arduino-workflow-router.
Sanitization: No sanitization or validation of the content within the JSON or Markdown files is described.

board-support