blackbaud
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSCREDENTIALS_UNSAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the vendor's
@useclawlink/clivianpxto perform authentication, action discovery, and CRM execution tasks.\n- [EXTERNAL_DOWNLOADS]: Fetches the@useclawlink/clipackage from the NPM registry as part of the integration setup and runtime execution.\n- [CREDENTIALS_UNSAFE]: The documentation identifies that authentication tokens and session keys are stored locally at~/.clawlink/credentials.jsonas part of the tool's standard operation.\n- [INDIRECT_PROMPT_INJECTION]: The skill uses natural language input to search for and describe Blackbaud actions.\n - Ingestion points: Natural language search strings in
npx @useclawlink/cli actions blackbaud "<what you want to do>"(SKILL.md).\n - Boundary markers: None explicitly defined in the CLI command syntax.\n
- Capability inventory: Execution of read and write actions within the Blackbaud CRM environment.\n
- Sanitization: Performed by the vendor CLI tool prior to API interaction.
Audit Metadata