The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill acts as a coordinator for gathering external information, creating a vulnerability surface where malicious instructions in researched data could influence agent behavior.
Ingestion points: External data collected from various sources during the research phases and user-supplied topic strings.
Boundary markers: The skill does not implement delimiters or 'ignore' instructions to isolate researched content from the execution context.
Capability inventory: The skill uses tools including Bash, Write, and Agent for task execution and file management.
Sanitization: No evidence of input validation or content filtering is provided in the instruction set.

research-pipeline