evernote-local-dev-loop

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill facilitates the creation of a development environment that ingests and processes data from the Evernote API, which is considered an untrusted external data source.
  • Ingestion points: The implementation guide shows external data entering the application through evernote.userStore.getUser() and evernote.noteStore.listNotebooks() in src/index.js and src/test-note.js.
  • Boundary markers: The instructions do not define specific delimiters or instructions to isolate the data retrieved from Evernote from the agent's prompt context.
  • Capability inventory: The skill environment is configured with Bash(npm:*), Bash(node:*), Write, and Edit tools, allowing for the execution of the generated application code.
  • Sanitization: The provided src/utils/enml.js utility includes a validateENML function that checks for a set of forbidden HTML elements, such as <script>, <form>, and <iframe>, providing a baseline level of sanitization against malicious content payloads.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 03:44 AM