The Agent Skills Directory

[COMMAND_EXECUTION]: The skill invokes specialized local Python scripts (registry-events.py, kg.py, pageviews.py, gdelt.py) to perform its primary functions of registry management and data retrieval. These operations are within the expected scope of the skill's purpose.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data via tool outputs and pasted content (ingestion points). It uses instructional warnings (boundary markers) to treat evidence as untrusted. The skill utilizes script execution (capability) to update the registry, while enforcing human-in-the-loop verification and schema validation (sanitization) for all persistent changes.
[SAFE]: The skill includes explicit operational guidance to protect user privacy, specifically advising on lawful bases for person-records and the exclusion of PII and credentials from system logs.

entity-registry