project-env-manager
Pass
Audited by Gen Agent Trust Hub on Apr 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through the 'project-env.yaml' file, which is an external data source. An attacker could embed malicious commands within the 'install.command' or 'verify.command' fields. Ingestion points: 'project-env.yaml' at the repository root. Boundary markers: The workflow includes a requirement for explicit user confirmation before any installation steps. Capability inventory: Execution of arbitrary shell commands as defined in the YAML contract. Sanitization: No automated validation of the provided command strings is performed.
- [COMMAND_EXECUTION]: The skill's core functionality relies on executing shell commands to audit and set up the environment. This capability presents a security risk if the configuration file is modified to include harmful commands.
Audit Metadata