The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to process and parse task descriptions, bodies, and response messages that may contain untrusted content from external sources (e.g., commit messages, PR descriptions, or user-provided data).
Ingestion points: Untrusted data enters the agent context through arguments in commands such as yy ledger create, yy ledger update, and yy ledger mark. The skill also automatically parses specific markup tags like [blocked_by] and [task_id] from within task body text.
Boundary markers: The instructions do not specify any delimiters or safety warnings to ignore instructions embedded within the task data.
Capability inventory: The agent has the capability to execute shell commands via the yy CLI tool and a local Python script (.juno_task/scripts/controller_resolver.py).
Sanitization: There is no evidence of sanitization, validation, or escaping of the external content before it is processed by the CLI or the agent.
[COMMAND_EXECUTION]: The skill facilitates the execution of the yy CLI tool for task management and references the execution of a project-local script .juno_task/scripts/controller_resolver.py for diagnostic purposes. These operations are intended parts of the skill's functionality for interacting with the ledger environment.

ledger-tasks-yylo