w-test
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [COMMAND_EXECUTION]: The skill identifies and runs test commands from
package.jsonusing local tools likepnpm. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from repository files and test logs which could contain malicious instructions.
- Ingestion points: Test command definitions in
package.jsonand captured execution logs. - Boundary markers: None present to delimit untrusted data from instructions.
- Capability inventory: Shell command execution via
pnpmand source code modification. - Sanitization: No sanitization is performed on external inputs.
- [NO_CODE]: The skill is comprised of instructional text and metadata with no executable code files included.
Audit Metadata