xl1-knowledge
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The analyzed skill consists of technical documentation, architectural patterns, and code examples for blockchain integration. It contains no executable scripts, hidden commands, or malicious instructions.
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion and processing of data from the XL1 blockchain and datalakes, which are external untrusted sources.
- Ingestion points: Blockchain blocks, transactions, and off-chain payloads are ingested through the SDK's gateway and datalake viewer interfaces, as documented in gateway.md and datalakes.md.
- Boundary markers: The skill does not define natural language boundary markers for data content, instead relying on programmatic SDK abstractions.
- Capability inventory: The skill enables network operations including broadcasting transactions to the chain and inserting payloads into datalakes via HTTP API calls.
- Sanitization: The skill mandates a 'Zod-First' type pattern for all protocol data structures, ensuring that every piece of ingested data is strictly validated against a schema before being used by the agent or application, significantly reducing the risk of processing malicious inputs.
Audit Metadata