discovering-connector-tools
Pass
Audited by Gen Agent Trust Hub on Aug 12, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONCREDENTIALS_UNSAFEPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references the installation of the author's official SDKs (scalekit-sdk-python, @scalekit-sdk/node) and well-known environment configuration libraries (python-dotenv, dotenv).
- [DATA_EXFILTRATION]: The skill communicates with the vendor's API (your-env.scalekit.com) to retrieve tool metadata, which is standard for its discovery purpose.
- [CREDENTIALS_UNSAFE]: The skill instructs users to utilize environment variables and .env files for secret management, adhering to secure development practices.
- [PROMPT_INJECTION]: The skill processes tool metadata from the Scalekit API, creating a potential surface for indirect prompt injection.
- Ingestion points: Data enters the agent context via get_tools (Python) and getTools (Node.js) calls in SKILL.md.
- Boundary markers: The skill does not implement delimiters or explicit instructions to ignore commands within the fetched metadata.
- Capability inventory: The skill's scope is restricted to metadata discovery; it explicitly lacks capabilities for tool execution, account authorization, or network write operations.
- Sanitization: No sanitization or content validation is provided for the retrieved metadata.
Audit Metadata