plugin-forge
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides utility Python scripts (
scripts/create_plugin.pyandscripts/bump_version.py) that perform local file system operations, such as creating plugin directory hierarchies and modifying manifest files (plugin.jsonandmarketplace.json). These are legitimate development tools triggered by specific user commands.\n- [INDIRECT_PROMPT_INJECTION]: The skill scripts ingest user-provided data (e.g., author name, description) and interpolate them directly into manifest and README files. This creates a surface for indirect prompt injection if these files are later processed by an agent without sanitization.\n - Ingestion points: CLI arguments in
scripts/create_plugin.py(author name, email, description, keywords).\n - Boundary markers: None present in the generated JSON or Markdown files.\n
- Capability inventory: Local file write operations in both scripts.\n
- Sanitization: No input validation or escaping is performed on the provided strings before writing to the file system.
Audit Metadata