yansu-agent-cli
Warn
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONPERSISTENCEINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill facilitates the execution of a local binary file using an absolute Windows path (C:\Users\Administrator.yansu-agent\bin\yansu.exe). This directs the agent to interact directly with the host filesystem and run executable code. While this is the primary purpose of the skill, absolute path execution of binaries is a sensitive operation.
- [PERSISTENCE]: The skill implements management of scheduled tasks via cron commands. It allows the agent to create (cron add), modify, and delete jobs that run at specified intervals (e.g., --schedule "every 5m"). This behavior enables automated, recurring execution of prompts or commands without direct user interaction in each instance.
- [INDIRECT_PROMPT_INJECTION]: The skill contains a vulnerability surface where untrusted data can be stored for future execution.
- Ingestion points: The --prompt parameter within the cron add command in SKILL.md.
- Boundary markers: There are no explicit delimiters or instructions provided to the agent to treat the prompt content as untrusted or to ignore embedded instructions.
- Capability inventory: The skill possesses the ability to synchronize project data (push/pull) and manage automated execution schedules.
- Sanitization: The instructions do not define any sanitization, escaping, or validation logic for the content passed into the --prompt argument.
Audit Metadata