wuji-cli-doctor
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions for the agent to use the
wujiCLI tool to perform diagnostics and manage device configurations. The agent is directed to use commands likewuji doctorfor status checks andwuji setfor parameter adjustments.- [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow for ingesting and processing data from external hardware devices. - Ingestion points: Real-time sensor data (approximately 100 frames) collected from devices such as
wuji_gloveandwuji_hand_2as described in SKILL.md. - Boundary markers: The skill instructs the use of
--jsonand--jsonlflags to ensure structured and delimited output from the CLI tool. - Capability inventory: The skill has the ability to execute shell commands, modify device parameters (
effort_limit,mit_params), and trigger physical device motion (thumb_opposition). - Sanitization: Ingested data consists of hardware sensor frames and predefined status verdicts (pass, fail, warn, skip) rather than untrusted natural language text.
Audit Metadata