plankton-code-quality
Warn
Audited by Gen Agent Trust Hub on Jul 4, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONREMOTE_CODE_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill requires users to clone a repository from an untrusted GitHub account (
https://github.com/alexfazio/plankton.git) to obtain the necessary scripts for operation. - [COMMAND_EXECUTION]: It configures the agent to automatically execute shell scripts, specifically
multi_linter.sh,protect_linter_configs.sh, andstop_config_guardian.sh, during standard tool-use phases (PreToolUse, PostToolUse, and Stop hooks). - [REMOTE_CODE_EXECUTION]: The skill documentation describes an automated "Delegate + Verify" phase that spawns
claudesubprocesses with permissions to modify project files. This allows the system to autonomously edit code based on the output of local linters, creating a dynamic execution chain influenced by the content of the files being processed.
Audit Metadata