The Agent Skills Directory

[SAFE]: Technical review of the skill shows it contains only instructions for the AI model. No suspicious commands, unauthorized file access, or network exfiltration attempts are present.
[PROMPT_INJECTION]: The skill is designed to process external research data, creating a surface for indirect prompt injection. This is evaluated as a functional aspect of the research synthesis task.
Ingestion points: Research files provided by the router (SKILL.md).
Boundary markers: None explicitly defined in the synthesis instructions.
Capability inventory: Read, Write, Edit, and Bash tools (SKILL.md).
Sanitization: The skill relies on the model's summarization capabilities without explicit data sanitization.

research