sui-jsonrpc-skill
Pass
Audited by Gen Agent Trust Hub on Mar 29, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill targets 'https://fullnode.mainnet.sui.io', which is a well-known and established public fullnode for the Sui blockchain.
- [COMMAND_EXECUTION]: The skill utilizes the 'uxc' utility to create command aliases and manage blockchain subscriptions, which is the core intended functionality of the skill.
- [SAFE]: The validation script 'scripts/validate.sh' is used for structural integrity checks and contains no malicious logic.
- [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection through the ingestion of external blockchain data. 1. Ingestion points: Untrusted data enters the context via 'sui_getObject' and 'suix_subscribeEvent' calls as documented in SKILL.md and usage-patterns.md. 2. Boundary markers: No explicit delimiters or safety instructions are provided to help the agent distinguish between data and embedded instructions. 3. Capability inventory: The 'uxc' tool has capabilities for network access and local file system writing (via the '--sink' parameter), which could be misused if the agent is manipulated. 4. Sanitization: There is no evidence of data sanitization or filtering of the content retrieved from the blockchain before it is presented to the agent.
Audit Metadata