The Agent Skills Directory

[COMMAND_EXECUTION]: The skill documents the use of the rect command-line interface to interact with a task board. This includes commands for listing, reading, and commenting on tickets, which are standard tool interactions within a restricted execution environment.
[INDIRECT_PROMPT_INJECTION]: The skill facilitates the processing of untrusted external data by reading ticket content from a remote board.
Ingestion points: External data enters the agent context via rect list, rect read, and rect search commands defined in SKILL.md.
Boundary markers: The instructions do not specify boundary markers or delimiters for the ingested ticket content.
Capability inventory: The agent can perform write operations to the board including rect add, rect comment, rect claim, and potentially rect close/reopen (if full-access is enabled).
Sanitization: No explicit sanitization or filtering of the ticket content is mentioned.

rectangular