nasiko-control-plane
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions to execute the
ecccommand for status checks and lifecycle management. The commands are constrained to specific subcommands (status,install,uninstall) with hardcoded flags like--jsonand--dry-runto ensure predictable behavior. - [EXTERNAL_DOWNLOADS]: The skill references downloading a CLI binary from
https://github.com/Nasiko-Labs/nasiko. This is mitigated by instructions requiring the pinning of a specific version (v0.1.0), SHA-256 digest verification, and a prohibition against executing unverified shell bootstrap scripts. - [INDIRECT_PROMPT_INJECTION]: The skill involves processing structured JSON output from tool commands, which is a potential surface for indirect prompt injection. However, the instructions mitigate this risk by requiring the agent to present dry-run data for user review before proceeding with installation or removal, ensuring a human-in-the-loop validation step.
Audit Metadata