custom-sounds
Warn
Audited by Gen Agent Trust Hub on Aug 11, 2026
Risk Level: MEDIUMPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill contains 'Self-Evolving Skill' instructions that command the agent to 'fix this file immediately' if instructions are incorrect or workarounds are needed. This provides a mechanism for the agent to persistently modify its own code and instructions, which could be abused to store malicious logic or bypass future constraints.
- [COMMAND_EXECUTION]: The skill executes a local TypeScript library located at
_lib/pushover_core.tsusing thebunruntime. The path is dynamically resolved using thecc-plugin-rootcommand. - [DATA_EXFILTRATION]: The shell commands explicitly unset proxy environment variables (
env -u HTTPS_PROXY -u HTTP_PROXY). While this is often done to ensure direct API communication, it also bypasses potential security monitoring or logging performed by network proxies.
Audit Metadata