skills/polya20/grill-me/triage/Gen Agent Trust Hub

triage

Pass

Audited by Gen Agent Trust Hub on Jun 15, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it ingests and acts upon untrusted data from an issue tracker.
  • Ingestion points: In SKILL.md, the 'Gather context' step specifically instructs the agent to read the full body and all comments of an issue, which are provided by external reporters.
  • Boundary markers: The instructions lack explicit boundary markers or 'ignore' directives to prevent the agent from following instructions embedded within the issue content itself.
  • Capability inventory: The skill possesses significant capabilities, including running shell commands for bug reproduction, writing to the .out-of-scope/ directory, and posting automated comments to the issue tracker (SKILL.md).
  • Sanitization: There is no mechanism described for sanitizing, filtering, or escaping the external content before it is interpolated into the agent's reasoning process.
  • [COMMAND_EXECUTION]: The bug reproduction workflow requires the agent to execute code or commands within the project environment.
  • Evidence: In SKILL.md, the 'Reproduce (bugs only)' section directs the agent to 'run tests or commands' based on 'the reporter's steps'. This creates a surface where a malicious reporter could craft reproduction steps that attempt to execute unauthorized commands if the agent follows the steps too literally or without sufficient isolation.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 15, 2026, 03:36 AM
Security Audit — agent-trust-hub — triage