Ticket arrives
A customer question comes in through Freshdesk (Zendesk next).
Cortex drafts support replies from your trusted docs and tickets, with citations your team can verify.
Connect your knowledge base, procedures, and support history. Cortex retrieves the right sources, drafts a reply, and shows exactly where the answer came from before your team sends it.
ticketHow do I reset my 2FA device?
retrievesecurity-guide.md §4.2 + 2 prior tickets
draftreply generated - citations attached
reviewawaiting agent approval before send
Citations Every draft links to its source
Review before send A human reviews before it sends
Full audit log Every generated answer is logged
EU-hosted Data stays in Europe
How it works
From ticket to verified answer, with a human in the loop.
Cortex never sends a reply on its own. It drafts. Your team reviews, verifies the sources, and sends.
Cortex retrieves
Cortex searches your trusted docs, tickets, and procedures for relevant context.
Cortex drafts a reply
A draft answer is generated from the retrieved sources, not from memory.
Citations attached
The draft links back to the exact document and passage it came from.
Agent reviews
A support agent reads the draft and the sources before doing anything else.
Agent sends
The agent edits if needed, then sends the reply from Freshdesk.
Logged for audit
The ticket, sources, draft, and outcome are logged for review and improvement.
About the canopy corporation
Why we built this.
Support teams answer the same questions over and over, searching through docs, old tickets, and procedures for every reply. General AI chatbots do not know your internal knowledge, and a confident wrong answer sent to a customer is worse than no answer at all.
We built Cortex to solve that: it retrieves the right passage from your trusted knowledge base, drafts a reply with citations attached, and puts a human in control before anything is sent. Every answer can be verified. Every action is logged.
We also build Canopy Host, a Git-based deployment platform - a smaller, secondary product from the same company. Both run on EU infrastructure, owned by your team.
Our main product
Cortex is what we build for.
Cortex helps support teams draft verified AI answers from trusted docs and tickets, with citations, audit logs, and human review before anything sends.
Cortex
AI support copilot
Connect Freshdesk, ingest your docs and old tickets, and start drafting cited replies. Cortex retrieves the right passage, drafts an answer, and keeps a human in control before it sends.
- Draft replies grounded in your docs and tickets
- Every answer cites its source
- Human review before anything sends
- Full audit log of every generated answer
- Scoped API keys, project isolation
- EU-hosted data
Also from Canopy · secondary product
Canopy Host
Git-based cloud deployment, launching Q3 2026. Connect a repo, get a running app.
Meet Cortex.
Built for support teams that cannot afford hallucinated answers. Cortex retrieves the right passage from your docs and tickets, drafts a reply, and cites its sources - so a human can verify before it ever reaches a customer.
Draft replies with citations
Every draft links back to the exact document and passage it came from. No hallucinated sources, no guesswork.
Knowledge bases
Upload docs, procedures, and past tickets, or connect external sources. Cortex chunks, embeds, and indexes everything automatically.
Human review before send
Cortex never sends a reply on its own. Every draft is queued for a person to check, edit, and approve first.
Dashboard control plane
Upload documents, test your assistant in the playground, inspect citations, manage API keys, and view usage.
Audit every answer
Conversation history, ingestion status, API key usage, and a full audit trail. Scoped to your project, always visible.
Strict isolation
Every document, embedding, and conversation is scoped to your project. No cross-customer data access, ever.
Fine-tuning (advanced)
For teams with specialised domain language. LoRA/QLoRA adapters, project-owned, never shared. Not required for most support workflows.
Built for support teams first.
The first Cortex workflow is a B2B support team using Freshdesk (Zendesk next): a ticket comes in, Cortex searches trusted docs and tickets, drafts an answer with citations, and an agent reviews it before it sends. Product and developer teams can use the same retrieval and API layer for internal knowledge and custom integrations.
The first packaged Cortex workflow
Symphony is the Freshdesk workflow powered by Cortex.
Symphony connects your Freshdesk workspace to Cortex. It reads incoming tickets, retrieves relevant context from your knowledge base, and surfaces a cited draft reply for your team to review.
It is not a separate product or a separate buyer journey - it is Cortex, packaged for the support use case. Freshdesk webhooks feed a Cortex assistant, your own knowledge bases provide the context, and everything stays inside your Cortex project boundary.
- AI-drafted replies grounded in your knowledge base
- Citations link back to the exact document used
- Runs fully within your Cortex project — no shared data
- Agent assignment and triage via Cortex API
How do I reset my 2FA device?
Hi, I've lost access to my authenticator app and can't log in. What are my options?
Cortex: suggested reply
Source: security-guide.md §4.2
Prove it during the pilot
Designed to reduce search time and speed up first response.
We do not claim results before you have measured them on your own tickets. The pilot exists to track these metrics on your data, so you decide with evidence, not a demo.
Search time per ticket
Track how long agents spend searching docs and old tickets before Cortex, and after.
First response time
Measure how quickly a draft is ready for review once a ticket arrives.
Draft acceptance rate
Track how often agents send a Cortex draft with little or no edit.
Tickets with a useful draft
Measure the share of tickets where Cortex produces a usable starting point.
Edit / reject rate
Track how often agents rewrite or discard a draft, and why.
Citation accuracy
Spot-check whether cited sources actually support the drafted answer.
Escalation reduction
Track whether fewer tickets need senior or specialist escalation.
Agent onboarding time
Measure how quickly a new support agent becomes productive with Cortex drafting from day one.
For developers
OpenAI-compatible.
Drop-in ready.
Cortex exposes an OpenAI-compatible chat completions endpoint. Point your existing code at your Cortex project and it just works, with retrieval-augmented answers from your own knowledge base.
- Scoped API keys per integration, never org-wide
- MCP server for native agent + tool integration
- Cortex CLI for scripting, CI pipelines, and local setup
- Webhooks, streaming responses, and conversation history
from openai import OpenAI
client = OpenAI(
base_url="https://api.cortex.canopy.pm/v1",
api_key="crtx_sk_...",
)
response = client.chat.completions.create(
model="cortex-assistant",
messages=[
{
"role": "user",
"content": "How do I reset 2FA?",
}
],
)
# Citations included in the response
print(response.choices[0].message.content)Also from Canopy · secondary product
Canopy Host: Git-based deployment.
Connect GitHub, GitLab, or Bitbucket. Push your code. Canopy Host detects your stack, builds a container, and deploys it with automatic HTTPS and custom domains. Launching Q3 2026.
Not the main story here
Cortex is our main product and commercial focus. Canopy Host is a smaller, separate product from the same company, for teams that also need somewhere to deploy their software.
Trust & security
Verify every answer. Keep humans in control.
Cortex is built so support teams can trust what it drafts: every answer cites its source, a human reviews before anything sends, and every generated answer is logged. Data is hosted on Hetzner infrastructure in European data centres by default. Provider API keys are encrypted at rest. Your knowledge base data is never used to train shared or platform-level models.
Human review before sending
Cortex drafts. Your team reviews, verifies the sources, and sends. Never the other way round.
Source citations
Every draft links back to the exact document and passage used, so you can verify it.
Project isolation
Every Cortex resource is scoped to your project boundary. No cross-customer data access.
Encrypted provider credentials
AES-GCM envelope encryption. Keys are never returned or logged.
Audit log included
Every generated answer, and who accessed what, when, and from which API key.
EU-hosted by default
Hetzner infrastructure in European data centres.
Pricing
Start with a defined Cortex support pilot.
A working pilot within 14 days, on your own Freshdesk tickets and docs. Canopy Host is a separate, secondary product launching Q3 2026 with a free Starter plan.
Cortex support pilot · Freshdesk first
Draft cited replies
your team can trust.
Connect Freshdesk, ingest your docs and old tickets, and start drafting cited replies. We promise a working pilot within 14 days, with pricing locked in for 90 days.
Setup from €750 · Monthly from €250/mo · See full pricing
Questions? hi@canopy.pm