The Agent Skills Directory

[Data Processing Surface]: The skill ingests data from external sources including pasted text, uploaded files, and connected platforms like knowledge bases and support tickets. This represents a potential surface for indirect prompt injection where instructions could be embedded in the data. Ingestion points: Gather Research Inputs section in SKILL.md. Boundary markers: No specific delimiters or instructions to ignore embedded commands are present. Capability inventory: The skill is limited to text generation and does not have the ability to execute commands or access the network. Sanitization: Input content is processed directly without explicit filtering.
[No Risky Operations]: The skill does not attempt to access sensitive system files, execute shell commands, or establish unauthorized network connections.

synthesize-research