create-cowork-plugin

Pass

Audited by Gen Agent Trust Hub on Mar 3, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes local system utilities (zip, cp) and an environment-specific validation tool (claude plugin validate) during the packaging phase to deliver the final plugin artifact to the outputs directory.
  • [SAFE]: The skill serves as a legitimate scaffolding engine for developers. All operations are transparent and consistent with the intended purpose of creating structured plugin components (Skills, Commands, Agents).
  • [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection by processing untrusted user input to generate instructions for the created plugin files. Ingestion points: User requirements via AskUserQuestion; Boundary markers: Absent in generated files; Capability inventory: Write, Edit, Bash; Sanitization: Absent.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 3, 2026, 06:42 PM