workflow-commits
Pass
Audited by Gen Agent Trust Hub on Apr 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill enables the agent to execute scripts derived from the project's package.json file. This is the core functionality and is managed through instructions to avoid generic commands.
- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it processes untrusted data from the package.json file to guide agent behavior.
- Ingestion points: package.json (relative path).
- Boundary markers: Absent; no instructions are provided to delimit or ignore instructions that may be embedded in the file content.
- Capability inventory: Execution of shell commands through project scripts.
- Sanitization: Absent; the skill does not include validation or sanitization of the input file contents.
Audit Metadata