lifi
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the external LI.FI REST API (e.g.,
/quote,/advanced/routes,/status) which is then processed by the agent to generate transaction summaries. This represents an ingestion point for potential indirect prompt injection attacks. - Ingestion points: API responses from
https://li.quest/v1andhttps://staging.li.quest/v1. - Boundary markers: Absent; there are no specific instructions provided to the agent to treat API responses as untrusted data or to ignore embedded instructions.
- Capability inventory: Network operations (GET/POST requests) and transaction signing flows through wallet adapters.
- Sanitization: The skill mitigates this risk by explicitly instructing the agent to provide a human-readable summary of all transaction details to the user for confirmation before signing.
- [EXTERNAL_DOWNLOADS]: The skill instructs the user/agent to install the official
@lifi/sdkpackage from the NPM registry and makes network calls to the LI.FI production and staging APIs. These resources are owned by the vendor and are required for the skill's functionality.
Audit Metadata