test-runner
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill constructs shell commands by interpolating a user-provided 'scope' string into templates for RSpec, Minitest, Jest, pytest, and Go. If the agent does not sanitize this input, it could lead to command injection (e.g., appending shell metacharacters like ';' or '&&' to the scope).
- [INDIRECT_PROMPT_INJECTION]: The skill reads and parses stdout/stderr from external test suites, which is untrusted data.
- Ingestion points: Output streams from test frameworks (SKILL.md).
- Boundary markers: Absent. The skill does not instruct the agent to use delimiters or treat the output as untrusted.
- Capability inventory: The skill has access to shell execution via various package managers and language runtimes (npx, bundle exec, etc.).
- Sanitization: Absent. The instructions parse the output directly for failure details without filtering.
- [DYNAMIC_EXECUTION]: The skill uses project file detection (e.g., checking for 'jest.config.js' or 'go.mod') to dynamically determine which shell commands to execute at runtime.
Audit Metadata