opendeploy-ops
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill leverages specific bash tools for infrastructure management. It explicitly discourages shell composition (using symbols like ';', '|', or '&&'), which ensures that every command is individually processed and validated by platform security hooks.
- [EXTERNAL_DOWNLOADS]: The skill manages the lifecycle of the '@opendeploydev/cli' tool via npm. These operations are directed toward well-known package registries and are consistent with the vendor's own verified tooling.
- [PROMPT_INJECTION]: Automated detection of concealment patterns is identified as a false positive. The instructions to avoid manual probes in favor of the vendor's 'preflight' utility and to redact sensitive tokens are legitimate security-focused operational constraints.
- [PROMPT_INJECTION]: Indirect Prompt Injection Analysis: (1) Ingestion points: Reads service logs ('services logs') and deployment logs ('deployments logs') in SKILL.md. (2) Boundary markers: None explicitly defined in the prompt text. (3) Capability inventory: Can perform service mutations ('restart', 'stop', 'start', 'rollback') in SKILL.md. (4) Sanitization: Instructs the agent to summarize log content for the user and requires explicit consent for any mutations.
Audit Metadata