sui-sdks
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and rely on external documentation files located within the project's dependency tree, which could be exploited if those packages are compromised.
- Ingestion points: The agent is directed to read
node_modules/@mysten/*/docs/llms-index.mdand related markdown files (referenced inllm-docs.mdandSKILL.md). - Boundary markers: There are no instructions for the agent to use specific delimiters or to treat the ingested documentation as untrusted content.
- Capability inventory: As a development-focused agent, it typically has permissions to read/write files, execute shell commands, and perform network operations.
- Sanitization: No sanitization or validation steps are defined for the content of the documentation files before they are processed by the agent.
Audit Metadata