researcher-hand-skill

Pass

Audited by Gen Agent Trust Hub on Apr 2, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a shell command python -c "import platform; print(platform.system())" during its initialization phase to detect the operating system environment.
  • [EXTERNAL_DOWNLOADS]: The agent uses the web_fetch tool to retrieve content from external websites found during the research process.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to how it handles untrusted data.
  • Ingestion points: Untrusted data enters the agent's context through the web_fetch tool when reading web pages from search results.
  • Boundary markers: The instructions do not include explicit boundary markers or warnings to the agent to ignore instructions or malicious payloads embedded within the retrieved web content.
  • Capability inventory: The skill has access to powerful tools including shell_exec, file_write, and file_read in the same execution context as the untrusted data processing.
  • Sanitization: There is no evidence of sanitization or filtering applied to the fetched content before it is processed by the agent for synthesis or reporting.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 2, 2026, 11:00 PM