The handbook, conversational
Employees ask in plain language; the agent answers from your current policies with the citation attached — PTO math, benefits windows, parental leave, all grounded.
AI for HR teams
Policy answers for everyone — confidential cases handled like it matters.
The HR pack answers policy questions from your handbook with citations, takes confidential case intake with strict data boundaries, and routes to the right HR partner — trust built in, not bolted on.
Connects to Workday · Bamboo HR · ServiceNow — plus anything with an API
Object Explorer
ontology ▸ v3 · published · object · property · action permissions
Day one
What hr teams see when they log in.
Not a blank canvas — a working role with default goals, allowed tools, and the guardrails already on.
Intake that keeps confidence
Sensitive reports flow through a confidential intake path: minimal data captured, compensation and health fields locked away from the model, routing straight to the named partner.
The right partner, first time
Cases classify and route by type and sensitivity — the employee tells the story once, and the HRBP receives it structured.
Skills invoked · governed by policy & licensing
knowledge.searchdata.summarizecompliance.check_policycomms.draft_response
Systems
WorkdayBamboo HRServiceNow
In the pack
The HR Agent ships ready to work.
Searches approved HR knowledge, summarizes supplied context, and drafts a policy-grounded response; it does not create or route cases. Installed from the library, wired to your systems, and run through the same 8-gate chain as everything on Cortex.
- Search approved HR knowledge
- Summarize supplied case context
- Check a proposed response against policy
- Draft a response for review
Governed for this role
Open with policy. Sealed with people data.
HR data is the sharpest test of property-level permissions — compensation, health, and case fields stay locked while the agent answers freely from published policy.
- Compensation and health fields locked deny-by-default
- Case files visible only to the assigned partner — agent included
- Every answer cites the policy version it came from
- Intake trail sealed — defensible if a case ever escalates
audit/verify
ok: true
verifyChain ▸ chained SHA-256 · signed receipts
FAQ
What hr leaders ask.
Can the agent see salaries or medical information?
No. Those properties are locked deny-by-default at the ontology level — a read returns a 403 before any model is invoked, no matter how a prompt is phrased.
What happens with a harassment report?
It enters the confidential intake path: minimal structured capture, immediate routing to the named HR partner, no broad visibility — and a sealed trail that protects everyone involved.
How do answers stay current with policy changes?
Answers ground in your knowledge base, so updating the policy updates the agent — and every answer cites the policy version used, so you can audit what was said when.
Keep exploring
Where to go next.
Agent Library
Every HR agent and skill, filterable across the 21,500+ item catalog.
All roles
Ten more enterprise role packs — from Service to Field Operations.
Agent Studio
Extend the pack or build your own role — no code, full governance.
Pricing
Individual, Team, and Business — governance included on every plan.
Put a governed HR to work this week.
Start a free trial, install the pack, and connect your systems — or bring us one workflow and we'll demo it end to end.