sub2api-opencode
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest external data (Sub2API domain and API keys) to generate configuration files and perform connectivity tests. While standard for this configuration task, it represents a potential surface for indirect prompt injection if the user-provided data is not validated. * Ingestion points: Step 1 in SKILL.md (Sub2API domain, API key). * Boundary markers: Absent. * Capability inventory: File writing (opencode.json) and network operations (curl commands). * Sanitization: No explicit sanitization or escaping logic is provided for the interpolated user values.
- [COMMAND_EXECUTION]: The skill provides template commands using curl for verifying connectivity to Sub2API endpoints and export for setting environment variables. These are standard diagnostic tools and secret management practices and do not involve unauthorized execution.
Audit Metadata