solana-vulnerability-scanner
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external Rust source code for Solana/Anchor programs, creating a vulnerability surface where malicious code comments or identifiers could attempt to influence the agent's audit behavior.
- Ingestion points: The agent reads program source files (primarily
lib.rs) and configuration files such asAnchor.tomlandCargo.tomlfrom the user's workspace. - Boundary markers: There are no explicit instructions to wrap ingested code in delimiters or use "ignore" prompts to prevent the agent from obeying instructions embedded in the analyzed code.
- Capability inventory: The skill has the capability to read files, search codebase using
rg, and execute local testing tools likeanchor testandsolana-test-validator. - Sanitization: The instructions do not define any sanitization, filtering, or validation steps for the content of the programs being scanned before they are processed by the agent.
Audit Metadata