duckduckgo-search

Pass

Audited by Gen Agent Trust Hub on May 14, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill installs the 'duckduckgo-search' package from the official PyPI registry via pip or uv. This is a standard and well-known library for its claimed purpose.\n- [COMMAND_EXECUTION]: The skill utilizes 'python -c' to execute scripts that interact with the search library and save results to local JSON files. These commands are necessary for the skill's functionality and are documented clearly.\n- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface because it ingests data from the public internet into the agent's context.\n
  • Ingestion points: Search result titles, snippets, and bodies are retrieved from DuckDuckGo and printed in SKILL.md.\n
  • Boundary markers: No explicit markers are used to delimit external content from system instructions.\n
  • Capability inventory: The skill utilizes shell execution (Bash) and file system writes as described in SKILL.md.\n
  • Sanitization: Web content is processed directly without filtering for potential instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
May 14, 2026, 12:14 AM
Security Audit — agent-trust-hub — duckduckgo-search