agent-issue-investigator

Pass

Audited by Gen Agent Trust Hub on Mar 5, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill defines access to a Bash tool, allowing the agent to execute shell commands to reproduce and trace bugs.
  • [EXTERNAL_DOWNLOADS]: The agent is equipped with WebFetch and WebSearch tools, enabling it to retrieve content from external websites for investigation context.
  • [PROMPT_INJECTION]: The skill presents an attack surface for indirect prompt injection (Category 8) as it processes untrusted bug reports and has access to high-privilege tools.
  • Ingestion points: Input bug reports, test failures, and incident descriptions as specified in SKILL.md.
  • Boundary markers: No specific delimiters are used to isolate untrusted user data from the agent's instructions.
  • Capability inventory: The agent has access to Bash, Write, Edit, and WebFetch tools.
  • Sanitization: There is no evidence of input validation or sanitization to prevent the agent from executing instructions embedded in the input reports.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 5, 2026, 01:13 AM