make-content-editable

Pass

Audited by Gen Agent Trust Hub on Mar 8, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection because it is designed to ingest and process untrusted user data from the local file system to drive code generation and modification tasks.\n
  • Ingestion points: The skill reads markdown files (content/**/*.md, .mdoc, .markdown) and Vue component files as described in Step 0 and Step 1 of SKILL.md.\n
  • Boundary markers: There are no explicit instructions or delimiters provided to the agent to treat content within these files as data rather than instructions, nor are there warnings to ignore embedded agent-steering text.\n
  • Capability inventory: The agent has the ability to read files, write/modify local project files, and execute specific MCP tools (mcp__nuxt-ui__get-component and mcp__nuxt-ui__get-component-metadata).\n
  • Sanitization: The skill instructions do not specify any validation or sanitization of the input file content before using it to generate the final MDC and Vue component code.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 8, 2026, 09:08 AM