Skills
skills/k-dense-ai/claude-scientific-skills/perplexity-search/Gen Agent Trust Hub

perplexity-search

Pass

Audited by Gen Agent Trust Hub on Mar 3, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: Indirect Prompt Injection risk. The skill ingests untrusted data from web search results via the Perplexity API.
  • Ingestion points: scripts/perplexity_search.py (processes responses from litellm.completion).
  • Boundary markers: Absent; the tool retrieves and displays raw output from the model.
  • Capability inventory: scripts/perplexity_search.py (has file-writing capabilities via the --output parameter).
  • Sanitization: Absent; the content is not sanitized or filtered before being presented to the user or saved to a file.
  • [EXTERNAL_DOWNLOADS]: The skill documentation guides the user to install the litellm package from the official Python Package Index (PyPI).
  • [COMMAND_EXECUTION]: Provides local CLI tools, scripts/perplexity_search.py and scripts/setup_env.py, for executing searches and managing environment configuration.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 3, 2026, 08:48 PM