MERIDIAN SUITE — CONTEXT PLATFORM FOR AI AGENTS

Your AI agents are only as smart as your context.

Meridian keeps your customers, brand voice, and consent in one governed layer — and gives your agents real products to act through: outreach that lands, content that sounds like you.

Start freeSee how it works

mcp.meridiansuite.ai · OAuth 2.1 + PKCE · works with Claude, ChatGPT, and your own agents

THE THESIS

AI is a commodity. Context is the differentiation.

The model your agent runs on is interchangeable. The relationships, voice, history, and consent it acts on are not. Meridian gives every product — and every external agent — one governed context layer to act from.

01

Two products live today. One context underneath.

Pick where the pain is sharpest. Both run on the same context layer, so the second product gets smarter the day you add it.

PIPELINE, RUN BY AGENTS

LIVE

Outreach OS

Eight specialised agents find the right people, enrich them across 18 data providers, write in your voice, and orchestrate email, LinkedIn, and WhatsApp. Nothing sends without your approval.

  • Waterfall enrichment — free sources first, paid providers only when needed
  • Reply intelligence that reads intent and drafts the follow-up
  • Approval gates and per-tenant budget caps on by default
Start freeLog in

ON-BRAND CONTENT AT MACHINE SPEED

LIVE

Content OS

A brand-voice engine scores every draft against your voice and catches drift before it ships. Twenty creative skills — Veo 3, Runway, HeyGen, ElevenLabs — plus AEO tooling so AI assistants recommend you.

  • Voice DNA scoring and drift detection on every artifact
  • Skill marketplace with a no-code skill builder
  • Provenance ledger on everything AI-generated
Start freeLog in

ON THE ROADMAP

IN BUILD

Data OS — The shared context layer — identity, consent, voice DNA. Bundled with every product.

PLANNED

Media OS — Paid media planning, creative production, and attribution.

PLANNED

Governance OS — Provenance, lineage, audit, and EU AI Act tooling.

PLANNED

Web Experience OS — Sites that agents can read and act on — llms.txt, structured APIs.

PLANNED

Engage OS — Agentic decisioning across every customer touchpoint.

PLANNED

Analytics OS — Cross-product attribution and agent-readable reporting.

02

Ask in a sentence. Approve. Ship.

Meridian is built to be operated by agents — yours or ours — with a human hand on the gate.

01

Connect

Point Claude, ChatGPT, or your own agent at mcp.meridiansuite.ai. OAuth 2.1, scoped tokens, and spending caps from the very first call.

02

Ask

“Find 50 fintech founders who read our whitepaper, draft outreach in our voice, schedule at their best send time.” One sentence is the whole interface.

03

Approve & ship

Every agent step lands on the Action Board as it runs. Approve a sample, ship the rest, keep the audit trail.

Meridian system map: AI agents connect through an MCP gateway to a governed context layer, which feeds the operating suite.
data_os.query_segments → outreach.find_contacts → data_os.get_contact_context → content.score_voice → outreach.draft_message → outreach.schedule_campaign
03

Every product is a namespace.

One MCP endpoint — mcp.meridiansuite.ai. Scoped tenant tokens, per-tenant budgets, kill switches at every level. Listed on Claude Connectors and ChatGPT Apps.

outreach.*

Prospecting, drafting, scheduling, reply handling.

find_contactsdraft_messageschedule_campaignget_reply_intelligence
content.*

Briefs, drafts, voice scoring, publishing.

create_briefgenerate_draftscore_voicepublish
data_os.*

The context bundle every other tool draws on.

get_contact_contextquery_segmentscheck_consentresolve_identity
meridian.*

Cross-product orchestration.

invoke_agentrun_workflowlist_skills
04

Guardrails first, then speed.

Agents that touch your customers need more than a prompt. These are defaults, not add-ons.

Bring your own keys

Your provider keys, your data agreements. Budget gates default to $10 a day per tenant, with kill switches at every level.

Provenance on everything

Every AI-generated artifact carries a provenance record — lineage, consent, audit trail. Built for the EU AI Act, not retrofitted.

Compliant by region

GDPR, CCPA, PDPA, LGPD, APPI, DPDP, PIPA. Consent and retention are first-class fields in the context layer, not an afterthought.

05

Start free. Grow into the platform.

Sign up on one product and bring your own keys — you pay providers directly, Meridian orchestrates. Add products as the context layer starts compounding; bundles and enterprise plans when you're ready.

Start with Outreach OSStart with Content OS

MERIDIANSUITE.AI

Give your agents context worth acting on.

Start with Outreach OSStart with Content OS

Or point your own agent at mcp.meridiansuite.ai