skills/inco-fhevm/skills/ctoken/Gen Agent Trust Hub

ctoken

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill provides an integration path for an AI agent to read data from an external REST API and subsequently perform blockchain transactions, creating a standard data ingestion surface.
  • Ingestion points: Data is ingested from the Inco Indexer REST API (e.g., api.ctoken.inco.org) to retrieve token flows, activity, and holdings as documented in references/indexer-api.md.
  • Boundary markers: There are no specific instructions or delimiters provided to guide the agent in distinguishing between data returned by the API and its own internal instructions.
  • Capability inventory: The skill provides write access to the blockchain through functions like deposit, withdraw, and confidentialSend via the CTokenClient in SKILL.md.
  • Sanitization: The skill relies on standard SDK type safety and does not describe additional sanitization for potential instructions embedded in transaction metadata.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 09:59 AM
Security Audit — agent-trust-hub — ctoken