Support · Colleague: Remy

Resolve requests before they become tickets

BeforeQuery answers up to 75% of employee and customer requests without a human touching them — grounded in your knowledge, actioned on your systems, approved where it matters.

Live now
Grounded answers

Every answer traces to a paragraph you can read

Remy never invents. It retrieves the passages that support the answer, shows them alongside the response, and refuses to answer when the corpus does not cover it. The reader sees the same evidence the model saw.
  • Source paragraph shown on every answer, one click away
  • Refusal is a first-class response — no confident guessing
  • Conflict detection surfaces contradictions before the reader hits them
grounded answer · sources
answer

Priya’s Okta lockout can be cleared via self-serve MFA challenge — no IT intervention required. If MFA fails, escalate to IT on-call.

cited from
01
IT · Okta account lockout policy
runbook · updated 4d ago
02
HR · Emergency access procedure
policy · updated 2w ago
03
Ticket #48211 · priya · resolved
past ticket · similar match
Approval routing

Reversible writes gated. Destructive writes stepped up.

Every action carries a risk tier. Read-only actions default to autopilot. Reversible writes route to the correct approver on your matrix. Destructive writes add a step-up MFA challenge on the reviewer before Remy will execute.
  • One matrix drives every playbook — no per-flow re-configuration
  • Delegation flows automatically when the reviewer is out
  • Every write is logged with the entities involved and the rollback path
approvals · pending 4
unlock okta account
priya.sharma · self-serve · read-only
denyapprove
refund $128.40
stripe · ord_A4b2c · manager approval
denyapprove
revoke prod IAM role
aws · role/analyst · step-up MFA
denyapprove
reset zoom SSO
kiran.mehta · self-serve
denyapprove
The Context Graph

Who is asking, what they own, what breaks if you touch it

Remy reads the graph before it decides how to respond. Personalisation without form fields. Access checks that catch the policy conflict before the write, not after the audit.
  • Ingested from your HRIS, IdP, and system-of-record for entitlements
  • Refreshed continuously — no snapshot drift between reorgs
  • Powers both retrieval and action — one graph, many playbooks
context · priya.sharma
priyateamon-calloktagithubworkday
Measured, not asserted

The Reports dashboard shows what changed

Auto-resolution rate, median loop time, approval backlog, coverage per topic — visible from day one, not at the quarterly business review. If Remy is not moving the numbers, you see it before your team does.
  • Weekly digest to Slack for the owning team lead
  • Drill-down from any metric to the source loops
  • Export to your BI stack — same numbers, your dashboards
reports · this week
Auto-resolved
78.4%
+4.1 pt
Avg. loop time
11.4s
-2.1s
Approvals pending
12
-3
Coverage
94%
+1.2 pt
monsun
The Remy method

Visibility. Reasoning. Action.

Every playbook Remy runs follows the same three-beat rhythm — one that puts every write behind evidence and every action behind policy.

01

Visibility

Read the full context — who is asking, what they own, what they can access, what breaks if we touch it. Nothing acts on incomplete signal.

02

Reasoning

Match the request to a grounded answer or a playbook. Cite the source paragraph, weigh the risk tier, and route to the correct approver where policy demands it.

03

Action

Execute the write on your systems with an audit-log id, an entity list, and a rollback path. Confirm the outcome with the requester in the same thread.

How a request flows

From ask to resolution in one loop

Every request travels the same six stages. What changes is how many stages your policies let Remy execute without waiting on a human.

01
Request

In Slack, email, widget, MCP, or the helpdesk. Same voice, same context.

Slack · Teams · Zendesk · Widget · MCP
02
Retrieve

Grounded lookup against your knowledge with source paragraphs held aside.

Knowledge Studio
03
Reason

Read the Context Graph — who, what they own, what breaks if we touch it.

Context Graph
04
Approve

Route to the correct approver per policy. Reversible writes gated, destructive step-up.

Approval matrix
05
Execute

Act on Okta, Stripe, Workday, GitHub, or your own systems. Logged and attributable.

Actions runtime
06
Verify

Confirm the outcome with the requester. Feed the result back into future decisions.

Feedback loop
Watch it work

One request. Twelve seconds. Zero tickets.

This is the actual shape of a Remy loop — abbreviated for the page. Every line is a verifiable, reversible operation with an audit-log id.

slack · #it-help
live
$priya @ #it-help:my okta got locked, can you unlock me?
·
remy read Priya Sharma from HRIS — active, on-call, in Berlin context.graph#c8a1
·
remy check policy: self-unlock with MFA challenge allowed policy.matrix
·
remy issue MFA challenge to enrolled TOTP device okta.mfa/challenge
·
remy unlock account after successful challenge audit#a4b2c · reversible
remypriya:You're unlocked. Try signing in again — if you're stuck, I'll page IT.
resolved · 11.4s · 0 human touches·traceable · reversible · policy-compliant

Sensitive writes swap the middle line for an approval-router step. Destructive writes add a step-up MFA challenge on the reviewer. Same skeleton, different gates.

Enterprise-grade by default

The controls your security team is going to ask about

SOC 2 Type II
audited annually
GDPR
EU data residency
HIPAA-ready
BAA available
SSO / SAML
Okta · Entra · Google
SCIM 2.0
auto-provision
Audit log
every write, traceable
Bring your key
AES-256-GCM at rest
Bring your model
Anthropic · OpenAI · self-host
Zero training
your data stays yours
Rollback
every write reversible
TLS 1.3
in transit
99.9% uptime
SLA on Enterprise

Frequently asked questions

Common questions about Agentic Support

Zendesk, Freshdesk, Front, and Intercom today, with Jira Service Management and Salesforce Service Cloud in Phase 2. See /integrations for the current catalogue and role badges.
You choose per playbook and per action. Read-only actions default to autopilot; writes default to require approval; destructive actions require step-up MFA on the reviewer.
Every answer shows the exact source paragraph that supported it, one click away. When the corpus does not contain an answer, the Assistant says so and escalates rather than guessing.
Median first playbook is live inside 45 days. The first read-only playbook can run in a single afternoon once the knowledge sources are connected.

Run support on BeforeQuery

Book a demo and see what Agentic Support does on your own data — usually within 45 days.