skills/sendaifun/skills/manifest/Gen Agent Trust Hub

manifest

Pass

Audited by Gen Agent Trust Hub on Sep 24, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references and encourages the installation of the official @bonasa-tech/manifest-sdk package from the NPM registry and interacts with the official Solana mainnet RPC endpoint.
  • [INDIRECT_PROMPT_INJECTION]: The skill implements logic to ingest and process live market data from the Solana blockchain, which is external, untrusted data.
  • Ingestion points: Market.loadFromAddress and ManifestClient.getClientReadOnly methods used in read-market.ts and manifest-setup.ts to fetch chain state.
  • Boundary markers: None explicitly present in the instructional code; data is processed directly via SDK methods.
  • Capability inventory: The skill provides capabilities to build and send blockchain transactions via sendAndConfirmTransaction and batchUpdateIx in place-order.ts, global-order.ts, and amm-style-batch.ts.
  • Sanitization: Not present; the skill treats blockchain-derived market data as standard API inputs for the DEX logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 24, 2026, 01:27 AM