The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local shell script validate-task-fields.sh during the task refinement and validation phases (Steps 2, 2.5, 3, and 3.5). While this is used for task field validation, the exact implementation of the script is not provided.
[PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it processes untrusted data from external messages and task descriptions to automate task creation and refinement. Ingestion points: Messages are ingested in Step 1; task titles, descriptions, and agent outputs are read in Steps 2, 2.5, and 3.5. Boundary markers: No explicit delimiters or warnings to ignore instructions within the ingested data are defined. Capability inventory: The skill can execute shell scripts (validate-task-fields.sh) and trigger other complex skills like planning-tasks and executing-tasks based on the processed data. Sanitization: No sanitization or validation of the content of the ingested messages or task fields is described before they are used to propose actions or update criteria.

running-daily-tasks