skills/formio/ai/formio-angular/Gen Agent Trust Hub

formio-angular

Pass

Audited by Gen Agent Trust Hub on Aug 17, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external, potentially untrusted data model definitions (template.md and template.json) to generate Angular application code, which represents an indirect prompt injection surface.
  • Ingestion points: Data enters the agent context through the planner artifact pair (template.md and template.json) referenced in SKILL.md, AUTH.md, and resources/SKILL.md.
  • Boundary markers: The skill explicitly instructs the agent to treat these artifacts as the strict "source of truth" and "never invent" resource models (SKILL.md).
  • Capability inventory: The skill possesses the ability to execute shell commands (npm install, npx skills add) and perform file system operations to scaffold and write code for the Angular workspace.
  • Sanitization: The instructions focus on mapping JSON properties directly to code templates; there is no mention of explicit sanitization or escaping of untrusted template content during the code generation process.
  • [UNVERIFIABLE_DEPENDENCIES_AND_REMOTE_CODE_EXECUTION]: The skill performs remote operations to install tools and dependencies required for Angular development.
  • Trusted Sources: It downloads the official Angular skills library from github.com/angular/skills using npx skills add to delegate workspace scaffolding to verified tooling (BOOTSTRAP.md).
  • Well-Known Services: It fetches version metadata from unpkg.com for @formio/angular, @formio/js, bootstrap, and bootstrap-icons to ensure framework compatibility (BOOTSTRAP.md).
  • Package Management: It performs standard npm install operations for essential Angular and Form.io SDK packages as part of the project bootstrap process.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 17, 2026, 07:00 PM
Security Audit — agent-trust-hub — formio-angular