openserv-agent-sdk
Pass
Audited by Gen Agent Trust Hub on Jul 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill follows established secure practices for credential management. It instructs users to store sensitive information like API keys and wallet private keys in .env files. The included 'provision' function programmatically handles these credentials to facilitate registration with the vendor's platform infrastructure.- [EXTERNAL_DOWNLOADS]: The skill utilizes official vendor packages (@openserv-labs/sdk, @openserv-labs/client) and well-known, industry-standard libraries (zod, dotenv, tsx). No suspicious or unknown third-party dependencies were identified.- [PROMPT_INJECTION]: The skill's examples (e.g., the 'analyzeText' capability in examples/capability-example.ts) demonstrate patterns for ingesting untrusted data through arguments and interpolating it into prompts. While this creates a surface for indirect prompt injection, it is a standard design pattern for AI agents and does not indicate malicious intent. (Evidence: Ingestion points: capability 'args'; Boundary markers: absent; Capability inventory: file read/write, task creation, and platform-delegated LLM calls; Sanitization: absent).
Audit Metadata