The Agent Skills Directory

[COMMAND_EXECUTION]: The skill relies on executing local Python scripts (e.g., scripts/things3_read.py, scripts/things3_write.py) to interact with the Things3 application. It also includes a direct Python command execution to verify existing areas.
[EXTERNAL_DOWNLOADS]: Requires the installation of the things.py library from the official Python Package Index (PyPI) to enable reading the Things3 SQLite database.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted data (task titles, notes, and project names) from the external Things3 database.
Ingestion points: scripts/things3_dashboard.py and scripts/things3_read.py read data from the local Things3 SQLite database.
Boundary markers: No explicit boundary markers or delimiters for processed data are defined in the instructions.
Capability inventory: The skill has the capability to execute shell commands and modify/delete data within the Things3 application via scripts/things3_write.py.
Sanitization: No sanitization or validation of the content retrieved from Things3 is mentioned in the skill instructions.

things3-manager