skills/garagon/nanostack/nano-run/Gen Agent Trust Hub

nano-run

Pass

Audited by Gen Agent Trust Hub on Sep 22, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes specific local utility scripts bundled with the Nanostack installation, including init-config.sh, init-stack.sh, init-project.sh, and save-setup-artifact.sh, to perform environment detection and project configuration.\n- [COMMAND_EXECUTION]: Telemetry lifecycle scripts (skill-preamble.sh and skill-finalize.sh) are executed at the beginning and end of the session. These scripts check for a NANOSTACK_NO_TELEMETRY environment variable or a local disable file before proceeding.\n- [INDIRECT_PROMPT_INJECTION]: The skill dynamically adjusts its behavior and instructions based on the contents of project files (such as package.json or go.mod) and host adapter configurations.\n
  • Ingestion points: Inspects project root directories, stack hint files, and host-specific adapter JSON files located in ~/.claude/skills/nanostack/adapters/.\n
  • Boundary markers: Uses specific output templates and follow-up actions defined in onboarding-contract.md to transition the user into the development lifecycle.\n
  • Capability inventory: Performs project detection, configuration file generation, and execution of bundled shell scripts.\n
  • Sanitization: Input validation is performed by the save-setup-artifact.sh script, which enforces schema constraints and enum values for configuration data before persistence.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 22, 2026, 11:56 PM