chainlink-cre-connect-skill
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill implements robust defensive measures, including mandatory event verification and explicit user confirmation for all on-chain operations.
- [SAFE]: Explicitly forbids the handling or requesting of private keys, seed phrases, or other sensitive authentication material.
- [INDIRECT_PROMPT_INJECTION]: Addresses potential ingestion of malicious content from external blockchain events and documentation. The skill mitigates this by instructing the agent to treat external data as untrusted, requiring local cryptographic verification of events in references/events.md, and enforcing boundary rules in SKILL.md to ignore instructions that weaken safety protocols.
Audit Metadata