The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data from external websites retrieved via the Perplexity API, creating a surface for potential indirect prompt injection attacks.
Ingestion points: External content is retrieved by the chat_completion and web_search functions in scripts/perplexity_api.py.
Boundary markers: The retrieved search snippets and AI responses are presented to the agent context without specific security delimiters or wrapping.
Capability inventory: The skill performs network operations to api.perplexity.ai and prints the resulting text.
Sanitization: No filtering or validation of the external web content is performed before it is added to the output.
[SAFE]: The skill does not contain any malicious code, obfuscated patterns, or persistence mechanisms. It follows secret management best practices by using environment variables for API keys and only imports standard Python modules.

perplexity-search