Your safeguarding evidence is in six different systems. Your audit form wants it on one page.
Most local authorities now expect early years settings to complete an annual safeguarding self-assessment — and the September 2025 EYFS framework raised the bar with a new safeguarding training criteria annex, prolonged absence tracking, and stricter safer recruitment documentation. The audit itself isn't hard. The hard part is pulling evidence from your DBS records, training log, incident reports, policy documents, and daily practice notes into a single coherent picture. Your safeguarding audit agent reads across all of those sources, maps what you have against what your local authority template asks for, and drafts a completed self-assessment with evidence cited — not just boxes ticked.
What Your Agent Actually Does
Your agent completes safeguarding self-assessments by reading your existing records — so the audit reflects what you actually do, not what you wish you did.
Completes your local authority's self-assessment template
Whether your LA uses the DfE model, their own bespoke form, or the LSCB framework — your agent is configured to the specific template your authority requires. It maps your evidence to their questions, not the other way around.
Reads across your safeguarding records
DBS single central register, safeguarding training log, incident and concern records, policy review dates, recruitment files, absence tracking — your agent pulls evidence from wherever it lives. No more copying and pasting between six different documents.
Flags compliance gaps, not just paperwork gaps
Missing a DBS renewal? Training lapsed for a key person? No record of how you handle prolonged absences under the new EYFS requirements? Your agent flags the gap and explains what's needed — not just that something's missing, but why it matters.
Tracks the September 2025 safeguarding changes
The new safeguarding training criteria annex, updated safer recruitment requirements, and prolonged absence procedures are all reflected in your agent's analysis. It knows what changed and what your audit needs to show.
Generates an action plan from audit findings
The self-assessment is only useful if you act on it. Your agent produces a prioritised action plan from the gaps it finds — what needs fixing first, who's responsible, and what evidence you'll need to close the gap.
The real numbers.
| Manager time completing safeguarding self-assessment | £800–£1,500/year |
| External safeguarding consultant or audit | £500–£1,500 |
| Time gathering evidence from multiple systems | £400–£800/year |
| Realistic annual cost | £1,700–£3,800 |
| Agent build (one-off, configured to your LA template) | £2,000–£3,500 |
| Monthly running costs (hosting + AI usage) | £60–£120/month |
| EYFS and LA template updates | Included in first year |
| Realistic first-year total | £2,720–£4,940 |
Safeguarding audits exist because they catch problems before children are at risk. But when the audit becomes a form-filling exercise — ticking boxes against a template without actually examining your practice — it defeats the point.
The settings that do best at inspection are the ones where the self-assessment is genuinely reflective. Your agent helps you get there by doing the evidence assembly so you can focus on the reflection — what's working, what isn't, and what you're going to do about it.
Good fit / not a fit.
This works brilliantly for:
- Any Ofsted-registered setting completing an annual safeguarding self-assessment
- Settings where the designated safeguarding lead also runs the room
- Multi-site nurseries that need consistent safeguarding audits across locations
- Settings that received safeguarding-related actions from their last inspection
This probably isn't for you if:
- You have a dedicated safeguarding officer with capacity to complete the audit thoroughly
- Your setting is very small with only 1–2 staff and a simple safeguarding profile
- Your local authority provides hands-on support that includes completing the audit with you
Handled like the sensitive data it is.
Children’s records — observations, safeguarding logs, SEND plans — are special category data under UK GDPR, and we treat them that way. Your agent runs on Claude via AWS Bedrock with an EU-only inference profile — meaning prompts and outputs never leave the EU, and are never used to train a model. Protected under Anthropic’s enterprise Business Associate Agreement.
Every run is logged for audit, every output is a draft your manager reviews and approves, and your data lives in a tenant isolated from every other customer. Parental consent verification is built into data workflows, and we’ll hand you a pre-filled DPIA template you can drop into your own records.
Need UK-only data residency? We offer an Azure UK South deployment as an enterprise add-on for customers with stricter procurement requirements. Full security details →
Let's talk.
We'll start with your local authority's self-assessment template and your current safeguarding records — where they live, what format they're in, and which bits take the longest. Usually a 15-minute conversation.
hello@nimblecroft.com