flaky-test-detector

Pass

Audited by Gen Agent Trust Hub on Mar 6, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted input from external test results (JSON) and source code files, which constitutes a surface for indirect prompt injection. (1) Ingestion points: Test result JSON files and various source code files scanned during static analysis. (2) Boundary markers: No delimiters or specific instructions are used to separate data from commands. (3) Capability inventory: Reading local file system and executing a local Python analysis script. (4) Sanitization: The skill uses standard JSON parsing but does not sanitize or filter content like test names or code comments before they are processed by the agent.
  • [COMMAND_EXECUTION]: The agent is instructed to execute a local Python script scripts/analyze_test_results.py to process test data. This script is a safe, internal tool and does not perform network operations, access sensitive files, or execute arbitrary system commands.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 6, 2026, 10:20 PM