solana-mcp-vybe
Pass
Audited by Gen Agent Trust Hub on Jul 15, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the user to configure their environment to fetch data and interface with a remote server at
https://mcp.vybenetwork.xyz. - [REMOTE_CODE_EXECUTION]: For Claude Desktop users, the skill utilizes
npx mcp-remoteto execute a proxy that connects to the remote Vybe MCP server. This allows remote logic to be integrated into the local agent context. - [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface (Category 8) because it ingests data from the Solana blockchain, such as transaction metadata and account information, which could be crafted by malicious actors to influence the agent's behavior.
- Ingestion points: Tools such as
solana_get_transaction,solana_get_token_metadata, andsolana_get_account_infofetch data from the public ledger. - Boundary markers: None provided; the skill does not instruct the agent to ignore instructions embedded in the blockchain data.
- Capability inventory: The skill primarily performs read operations on the blockchain and does not appear to expose file-write or subprocess capabilities directly.
- Sanitization: There is no evidence of filtering or escaping of content retrieved from the blockchain before it is presented to the LLM.
Audit Metadata