scaffold-plugin

Pass

Audited by Gen Agent Trust Hub on Jun 23, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill's primary purpose is scaffolding project files, which is a common and legitimate development task. While it uses file modification tools, its scope is restricted to the current repository.
  • [PROMPT_INJECTION]: The skill ingests user-provided strings (plugin name and description) and interpolates them into repository files, creating an indirect prompt injection surface. This is a common pattern in scaffolding tools.
  • Ingestion points: User input gathered during the 'Gather Plugin Information' phase (SKILL.md).
  • Boundary markers: Absent. User strings are interpolated directly into JSON and Markdown templates without delimiters or warnings.
  • Capability inventory: The skill uses 'Write' and 'Edit' tools to modify the local file system.
  • Sanitization: Absent. No validation or escaping is performed on the user strings before writing.
  • [SAFE]: Use of the Bash tool is limited to creating git commits for the scaffolded files as documented in the workflow.
  • [SAFE]: All external URLs and identifiers (e.g., github.com/TrogonStack) are consistent with the identified author and project vendor.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 23, 2026, 10:51 AM
Security Audit — agent-trust-hub — scaffold-plugin