unit-testing-test-generate

Warn

Audited by Gen Agent Trust Hub on Apr 1, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The CoverageAnalyzer class in SKILL.md contains a method analyze_coverage that passes the variable test_command to subprocess.run. This pattern presents a command injection risk if the command string or its arguments are derived from untrusted input, allowing for the execution of arbitrary shell commands.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it parses and processes untrusted data from source code files to generate tests. (1) Ingestion points: The _analyze_python method in SKILL.md reads content from external file paths via the open() function. (2) Boundary markers: The skill does not use delimiters or provide instructions to the agent to ignore instructions embedded within the analyzed source code (e.g., in docstrings). (3) Capability inventory: The skill possesses file system read access and the ability to execute subprocesses via the CoverageAnalyzer class. (4) Sanitization: No sanitization or validation is performed on the data extracted from source files (such as function metadata or docstrings) before it is included in the context for test generation.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Apr 1, 2026, 03:42 AM
Security Audit — agent-trust-hub — unit-testing-test-generate