picoclaw-ai-assistant

Pass

Audited by Gen Agent Trust Hub on Mar 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches precompiled binaries and source code from the Sipeed official GitHub repository. Sipeed is a recognized vendor in the single-board computer and hardware industry.
  • [COMMAND_EXECUTION]: The installation process involves executing downloaded binaries, running compilation commands such as make and go build, and managing system containers with Docker. These are routine operations for a technical utility of this nature.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection risks because it processes untrusted content retrieved from web search tools. * Ingestion points: Data returned from the Tavily and Brave Search APIs. * Boundary markers: No specific delimiters or safety instructions are defined in the provided CLI configurations. * Capability inventory: The agent has iterative tool-use capabilities and access to a local workspace for file operations. * Sanitization: Verification of input sanitization is not possible as the core logic is contained within the application binary.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 17, 2026, 07:31 AM