Note: This article is updated weekly throughout the month as new SafeBase features are released. All updates are also published on our Product Updates site.
Table of Contents
SafeBase Chrome Extension – Portal‑Agnostic Parsing
What is Portal-Agnostic Parsing?
Portal-Agnostic Parsing unlocks support for all previously unsupported TPRM portals, enabling AI-assisted questionnaire completion regardless of portal design or format. It works by extracting questions directly from the HTML in the browser. It will extract only what is visible; however the customer has the ability to navigate through to subsequent pages and sections in the questionnaire and click to parse all questions.
Once imported into SafeBase, the questions are answered and made available directly in the Chrome Extension so that the user only needs to copy and paste the answers into the portal. Autofill is not available on unsupported portals.
Solution Highlights
Portal-agnostic question extraction from any third-party questionnaire or TPRM portal, no portal-specific build required.
Side panel questionnaire rendering inside the Chrome Extension, enabling structured review of extracted questions.
AI-Assistance for answering, allowing users to generate suggested responses and copy them into the portal efficiently.
Content Collections (Early Access)
Availability: Content Collections is currently available through Early Access.
Why this matters
When customers run AI Questionnaire Assistance (AIQA), the system pulls content from the entire Trust Library, filtered by the selected product.
For organizations managing complex product lines or multiple deployment models, this can sometimes result in AIQA responses that include content that isn’t relevant to the questionnaire being answered.
Content Collections resolves this by allowing users to direct AIQA to a specific subset of Trust Library content, helping generate more precise and relevant answers.
How it works
Curated content collections:
Admins can create collections—curated subsets of their Trust Library—by filtering on content type (documents, knowledge base entries, Trust Center items) and tags.
Scoped AIQA runs:
When running AIQA, users select a content collection to restrict which content is used to generate answers. Only one collection can be selected per AIQA run.
Trust Library Search integration:
Content collections also work with Trust Library Search and Ask AI, allowing users to filter search results to a specific collection.
Built-in testing:
Before creating and saving a content collection, users can test the collection to verify it returns the correct content before applying it to a questionnaire.
Non-destructive editing:
Edits to a collection only affect future AIQA runs. Past questionnaire results are never retroactively changed.
To learn more about this change and how it may impact your organization, refer to Trust Library - Content Collections.