The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local Python scripts to manage board operations.
Evidence: python {skill_path}/scripts/board_manager.py is called for listing channels, reading posts, and publishing messages.
[INDIRECT_PROMPT_INJECTION]: The skill reads data from an external communication hub that can be populated by other agents or users, creating a surface for indirect prompt injection.
Ingestion points: The --read command in SKILL.md fetches markdown content from board channels and injects it into the agent's context.
Boundary markers: Absent. There are no instructions or delimiters provided to help the agent distinguish between its own system instructions and the content of the board posts.
Capability inventory: The skill uses board_manager.py to perform file system operations (read/write) on the message board storage.
Sanitization: Absent. The instructions do not describe any process for filtering or escaping instructions that might be embedded within the markdown posts.

board