pico-env-doctor
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements strong security boundaries, including requiring explicit user authorization for repairs, advising against printing credentials during environment inspections, and instructing the agent to honor host-level project trust requirements.
- [EXTERNAL_DOWNLOADS]: The skill provides instructions for the installation and maintenance of the
@picoxr/pico-clipackage from the public npm registry. These operations are standard and directly related to the skill's primary purpose of managing the PICO Spatial toolchain. - [COMMAND_EXECUTION]: The skill utilizes a set of diagnostic commands (e.g.,
node --version,pico-cli doctor) to assess the environment's health. It also defines specific setup and update commands that are executed only when the user's task or explicit request authorizes environment mutation. - [INDIRECT_PROMPT_INJECTION]: The skill identifies an ingestion surface where it processes data from tool outputs (e.g.,
pico-cli doctor --format json). This behavior is consistent with the skill's role as a diagnostic 'doctor' utility, and the risk is mitigated by explicit instructions to re-verify findings after repairs.
Audit Metadata