limitr-network
Pass
Audited by Gen Agent Trust Hub on Nov 19, 2026
Risk Level: SAFEDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [DYNAMIC_EXECUTION]: The protocol allows providers to define 'logic' in Stof syntax that is executed on the consumer's side. This is a documented feature described as having restricted access to the host environment, functioning as a 'sealed' and time-limited sandbox.
- [INDIRECT_PROMPT_INJECTION]: By consuming external tool definitions and data from service providers, the skill creates a surface where an agent could potentially be influenced by untrusted external content, a risk common to any external data integration.
- [EXTERNAL_DOWNLOADS]: The skill utilizes the vendor's official SDKs, '@formata/limitr' for Node.js and 'limitr' for Python, to implement the network protocol functionality.
Audit Metadata