> For the complete documentation index, see [llms.txt](https://docs.sprinto.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.sprinto.com/product-updates/august-2026.md).
# August 2026
Headlined by a rebuilt Security Questionnaire answer-generation engine and AI Nudges surfacing gaps across the product, plus Security Questionnaire answer style customization, live answer progress, and the new Monitoring Details Page beta. Here’s what’s new.
## Security Questionnaire: Answer Style Customization
You can now tell Sprinto AI how you want your Security Questionnaire answers written — tone, length, and formatting — instead of accepting one default style for every answer.
### What’s new
* **Answer style presets:** From the Add Questionnaire drawer, choose one of three starting presets — Concise, Bullet points, or Formal tone — and each one pre-fills an editable instruction box, so you can fine-tune it further or use it as-is.
* **Fully custom instructions:** Prefer to write your own? Skip the presets entirely and enter a free-text instruction of up to 500 characters.
* **Applies consistently:** The instruction is saved per questionnaire and applied every time Sprinto AI generates or regenerates that questionnaire’s answers, including via CSV/file import.
* **Guardrails stay intact:** Your instruction can shape phrasing, tone, and formatting, but it cannot change the underlying Yes/No/N/A answer determination, cannot cause the AI to fabricate facts, and won’t change the answer’s output language.
### Why it matters
* Lets your team match your organisation’s existing questionnaire voice (e.g. formal for enterprise RFPs, concise for high-volume vendor forms) without manually rewriting every AI answer.
* Reduces post-generation editing time on questionnaires where a customer or prospect has a specific tone requirement.
**Availability:** Globally Available
## Security Questionnaire: Live Answer Progress
Security Questionnaires no longer lock you out while Sprinto AI generates answers — you can open the questionnaire immediately and watch answers populate as they complete.
### What’s new
* **Open while generating:** Enter a questionnaire as soon as generation starts, instead of waiting for it to finish.
* **Progress banner:** A live “X of Y answered” banner shows while generation is in progress; multi-sheet questionnaires show a progress count per sheet tab.
* **Answers fill in incrementally:** Sprinto AI processes questions in batches (a smaller first batch, so early answers show up faster) and reveals each batch as it completes.
* **In-flight rows are protected:** Rows still being generated are visually locked with a tooltip, and are excluded from bulk actions, inline editing, and export until their answer is ready.
### Why it matters
* No more waiting on a blocked “Generating…” screen for large questionnaires — start reviewing completed answers while the rest finish.
* The progress banner and per-sheet counts give a clear sense of how much longer generation will take.
**Availability:** This is an on-demand feature. Raise a request to get this feature enabled for your organisation.
## Security Questionnaire: Rebuilt Answer-Generation Engine
Sprinto AI’s Security Questionnaire answer generation has been rebuilt on a new multi-stage retrieval-and reasoning pipeline, aimed at producing better-grounded, more accurate answers from your knowledge base.
### What’s new
* **Relevance check first:** Before searching for evidence, the AI checks whether a question even applies to your business (e.g. skips hazardous-materials questions for a SaaS company) instead of guessing an answer.
* **Broader, smarter search:** Each question is expanded into multiple phrasings (including a “what would the ideal answer look like” variant) and searched in parallel against your knowledge base, then reranked in two passes to surface the most relevant evidence.
* **Separated reasoning steps:** One pass extracts only verbatim facts from the retrieved evidence, a second decides the answer direction (Yes / No / Not Applicable / Not Evidenced), and a third writes the final answer — with an automatic grounding check that can trigger one retry if the answer isn’t properly backed by its sources.
* **Reuses prior answers:** Questions that closely match one already answered elsewhere in your knowledge base can be answered directly from that match, skipping the full pipeline.
* **Source tracking:** Every generated answer records exactly which knowledge-base documents it was derived from.
### Why it matters
* Reduces hallucinated or off-topic answers by having the AI recognise when a question doesn’t apply, rather than forcing a guess.
* Better retrieval and a grounding check mean answers are more likely to reflect what’s actually in your knowledge base, not a plausible-sounding fabrication.
* Internal evaluation on a 30-question benchmark showed strong groundedness (faithfulness score 0.995) with only 1 false positive.
**Availability:** Live in production for orgs on the new pipeline; some orgs may still be on the previous answer generation flow as fallback.
## AI Nudges Across the Product
Sprinto AI now proactively points out gaps and offers to help fix them in several places across the product — not just inside the chat/agent experience.
### What’s new
* **Access Control — automated background matching:** Sprinto AI runs on a recurring schedule and automatically suggests matches between unmapped access-control users (Okta, GSuite, GitHub, etc.) and the correct staff member, shown inline with a plain-English reason. Approve or dismiss individually or in bulk.
* **Contextual “Review with AI” banners in six other areas:** vendor profiles, vendor risk scoring, vendor document requests, submitted vendor questionnaires, staff role descriptions (missing job descriptions), staff non-human-identity review, risk item scoring, and policy framework-update inconsistencies. Each banner appears when Sprinto detects a relevant gap and offers a one-click way to review or generate the fix with AI.
### Why it matters
* The Access Control nudge removes a genuinely tedious manual-matching task entirely — no more line-by-line reconciling of usernames to staff records.
* The banner nudges surface AI help exactly where and when it’s relevant, instead of requiring users to know an AI action exists and go find it.
## Monitoring Details Page (MDP) — Beta
A new in-place “Monitor details” drawer experience is rolling out as an opt-in beta, replacing the standalone check details page for supported monitoring areas.
### What’s new
* **In-place drawer, not a separate page:** Clicking a monitor/check now opens a drawer showing the entity table and monitor configuration in place, instead of navigating away from the list or Ops Dashboard.
* **Dedicated Reviews tab:** Management-review monitor types get their own Reviews tab inside the drawer.
* **Opt-in beta banner:** A “Switch to the new monitoring details experience” banner lets users try the new drawer ahead of general availability, with a built-in feedback / switch-back option if they want to revert.
### Why it matters
* Keeps users in context (list view or Ops Dashboard) instead of bouncing to a separate page and back for every monitor they check.
* The dedicated Reviews tab makes management-review monitors easier to action without hunting through a generic layout.
**Availability:** Beta, opt-in. Production rollout of the underlying capability is currently scoped to the People monitoring area only.
## Quick Updates
Smaller improvements and fixes shipped this month:
* **Policy Management:** SharePoint integration now supports multiple distinct SharePoint sites per org, not just multiple folders within one site.
* **Access:** Deactivation and last-working date are now shown in the staff-access and revoked-accounts tables.
* **Staff Devices:** The include/exclude control for employee and contractor devices — previously Dr. Sprinto only — now extends to third-party MDMs.
## Wrapping Up
That’s August. As always, if you’d like a deeper walkthrough of any of these, reach out to your CSM or drop a note in our support channel.
---
# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.
## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.
Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:
```
GET https://docs.sprinto.com/product-updates/august-2026.md?ask=&goal=
```
`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.
Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.