sdk-assistant-agent

Pass

Audited by Gen Agent Trust Hub on Mar 4, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the Bash tool to perform essential development tasks such as building firmware (build.sh), burning code to hardware (cskburn), and managing serial connections (picocom, fuser). These commands are standard for embedded development within the target SDK.
  • [SAFE]: The skill includes a self-evolution mechanism in evolution/SKILL.md that allows it to update its own instruction files and knowledge base using the Edit tool. This is a documented feature designed to maintain the agent's effectiveness as the SDK evolves.
  • [SAFE]: The skill processes potentially untrusted data from the local codebase and official documentation (docs2.listenai.com). Ingestion points: Local driver, component, and sample files; online documentation from docs2.listenai.com. Boundary markers: Relies on structured routing and specific tool execution flows. Capability inventory: Extensive use of Bash for system operations and Edit/Write for file system modifications. Sanitization: No explicit sanitization of ingested content before use in prompts is documented.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 4, 2026, 03:43 AM