Hedera Hook Creation
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides legitimate technical documentation and examples for extending the Hedera Agent Kit. It utilizes official libraries from the '@hashgraph' organization, which is a well-known and trusted entity. The provided code examples, such as the timing hook, demonstrate standard software engineering patterns for monitoring and session state management.
- [INDIRECT_PROMPT_INJECTION]: The skill provides patterns for processing tool input data which could originate from untrusted sources, creating a potential surface for indirect prompt injection.
- Ingestion points: Tool parameters are ingested via the 'rawParams' and 'normalisedParams' objects within hook methods defined in 'SKILL.md'.
- Boundary markers: The documentation does not specify the use of delimiters or 'ignore' instructions for processed tool data.
- Capability inventory: Hooks have access to the 'Context' object and can perform side effects like logging ('console.log') or writing to Hedera Consensus Service (HCS) topics via built-in hooks referenced in 'references/hook-interface.md'.
- Sanitization: The provided examples demonstrate 'JSON.stringify' for logging but do not include explicit validation or sanitization of the parameter content.
Audit Metadata