plugin
Pass
Audited by Gen Agent Trust Hub on Nov 3, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes shell commands including
curl,jq, andcatto communicate with a local language server service atlocalhost:5387. These commands are used to inventory installed plugins, update configuration states, and manage local plugin files. - [SAFE]: All network operations are restricted to the local host (
localhost), and the skill manages credentials (CSRF tokens) through standard environment variables rather than hardcoding or exfiltrating them. - [INDIRECT_PROMPT_INJECTION]: The skill explicitly identifies the risk of context flooding and potential instruction injection when reading plugin metadata. It mandates the use of
jqto filter RPC responses, ensuring only necessary data fields are processed and preventing the ingestion of unfiltered external instruction files.
Audit Metadata