dev-workflow
Fail
Audited by Gen Agent Trust Hub on Aug 4, 2026
Risk Level: HIGHCREDENTIALS_UNSAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [CREDENTIALS_UNSAFE]: The skill contains hardcoded default credentials for bootstrap admin accounts in staging and local development environments.
- Evidence: 'Password: Admin@staging2026' for staging in SKILL.md.
- Evidence: 'Password: ChangeMe!2026' for local admin in SKILL.md.
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands via make using user-provided parameters, which could lead to command injection if the inputs are not sanitized by the calling agent.
- Evidence: 'make parallel-init TASK=feature-name' and 'make parallel-sync SOURCES="feat1 feat2"' in SKILL.md.
- [PROMPT_INJECTION]: The skill's parallel workflow commands provide an ingestion point for untrusted data that is used in high-privilege shell operations without boundary markers.
- Ingestion points: TASK and SOURCES arguments in SKILL.md.
- Boundary markers: None present.
- Capability inventory: Execution of arbitrary shell commands via make.
- Sanitization: No sanitization or validation of the task/source strings is described.
Recommendations
- AI detected serious security threats
Audit Metadata