plugin-creator
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [DYNAMIC_EXECUTION]: The skill executes bundled JavaScript files (e.g.,
scripts/create-basic-plugin.mjs,scripts/upsert-dev-marketplace.mjs) using the Node.js runtime to automate the creation of plugin scaffolds and the updating of marketplace JSON files. - [COMMAND_EXECUTION]: The skill instructions allow for the invocation of external CLI tools, such as
zcode plugins validate, to perform manifest verification if the developer environment supports it. - [INDIRECT_PROMPT_INJECTION]: The skill operates as a code generator that ingests user requirements to implement plugin logic, creating a vulnerability surface where malicious instructions could be embedded in the generated output.
- Ingestion points: User-provided choices for purpose, name, and components are used to generate plugin implementation files (SKILL.md, Section 1).
- Boundary markers: The skill does not explicitly instruct the agent to use delimiters or ignore instructions found within the user-provided plugin requirements.
- Capability inventory: The skill can execute Node.js scripts and perform file writes to create directories, plugin manifests, and marketplace catalogs.
- Sanitization: No sanitization or validation of user-provided content is mentioned before it is interpolated into the generated code templates.
Audit Metadata