The Agent Skills Directory

[PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data from external sources such as interview transcripts, survey results, and online community posts (Reddit, G2, etc.). This creates an indirect prompt injection surface where malicious instructions in the source material could attempt to influence the agent. Ingestion points: external research assets and community forums. Boundary markers: not specified in the instructions. Capability inventory: the skill lacks high-risk execution capabilities such as shell access or dynamic code evaluation. Sanitization: no explicit validation or sanitization of input data is defined.
[SAFE]: No malicious patterns such as obfuscation, hardcoded credentials, or remote code execution were found. All external references point to legitimate research tools and well-known platforms.

customer-research