test-review

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data by reading source and test files from user-specified directories.
  • Ingestion points: Sub-agents read all source and test files within the provided module and test paths in Phase 2.
  • Boundary markers: No explicit delimiters or warnings to ignore embedded instructions are included in the sub-agent prompts when reading code content.
  • Capability inventory: The skill uses Glob tools for file discovery and the Task tool to spawn sub-agents for reading and summarizing code. The main agent subsequently processes these summaries to generate its report.
  • Sanitization: No sanitization or filtering of file content is performed before it is processed by the AI models.
  • [COMMAND_EXECUTION]: The skill executes the cat command to read its own internal reference files, such as testing-standards.md and audit-workflow.md, to calibrate the audit process.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 12:41 PM
Security Audit — agent-trust-hub — test-review