cm
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill downloads the Bun runtime installer and its own installation and update scripts. These downloads are performed from well-known (bun.sh) and vendor-owned (GitHub) sources.
- [REMOTE_CODE_EXECUTION]: Installation and update commands utilize the
curl | bashpattern to fetch and execute scripts from the vendor's official GitHub repository. This is the intended distribution and update mechanism for the tool. - [COMMAND_EXECUTION]: The tool executes various CLI commands including
gitfor repository integration,sshfor remote session search, and a separatecassbinary for indexing. It implements validation for SSH targets to prevent common command injection vectors. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process AI agent conversation logs. While this creates a vulnerability surface for indirect prompt injection, the system includes a dedicated sanitization module to redact sensitive information like API keys and passwords before the data is processed or sent to an LLM.
Audit Metadata