test-analyze

Pass

Audited by Gen Agent Trust Hub on Feb 27, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted data from external files (XML, JSON, HTML) specified in the results path. This creates a surface for indirect prompt injection where malicious content inside a test report could influence agent behavior.
  • Ingestion points: The prompt instructs the agent to read and extract data from the path provided in {{ARG1}} using Read, Glob, and Grep tools.
  • Boundary markers: The instructions do not define delimiters or provide specific warnings to the model to ignore embedded instructions within the ingested test data.
  • Capability inventory: The skill possesses significant capabilities including systematic planning via mcp__pal__planner and task delegation to multiple subagents (security-audit, code-refactoring, etc.) via the Task tool.
  • Sanitization: There is no evidence of content sanitization or validation before the data is passed to the planning model or subagents.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 27, 2026, 07:26 PM