The team distribution layer across AI platforms
Your team's AI,
now distributed.
Context, memory, and shared skills — on every IDE your team uses. No walled gardens.
Works with Claude Code · Cursor · Codex · Copilot
Works with
One source of truth. Every surface your team already uses.
How it works
Three pillars. One team layer.
Context
Auto-loaded on every new session.
- ·Project state
- ·Main-branch truth
- ·Open decisions
Memory
Recall the decision, not the doc.
- ·Prior sessions
- ·PR threads
- ·Design docs
- ·Text-to-SQL
Registry
Publish once. Install anywhere.
- ·Skills
- ·Agents
- ·Connectors
- ·Playbooks
- ·Versioned & approved
A day in a Maiife team
Three moments. One shared surface.
Morning context. Afternoon recall. Evening publish. Every IDE on the team, one source of truth.
Claude Code opens.
Maiife auto-loads the team's context before the first message.
Maiife context loaded
14 facts · 2 new decisions since yesterday · last sync 3m ago
Stuck on an old design call.
Ask memory, not a teammate.
> /maiife recall "auth provider decision"
→ Decision (PR #412, Mar 14): dropped Auth0, moved to Supabase Auth. Reason: cost + self-host path. Owner: @sakthi.
Built something reusable.
One command and your team has it tomorrow.
$ maiife publish ./skills/triage-incident
✓ Published skill triage-incident@1.0.0 ✓ Live in 4 workspaces · propagation 2m 41s
Plays with what you already have
We don't replace your CLAUDE.md. We enrich it.
Your tool-native files stay put. Maiife adds the team layer on top — main-branch truth, open decisions, shared skills.
Configurable per workspace. Audit-logged.
Default
Enrich
Maiife merges team context into your tool-native file at session start. Main wins on conflict.
One platform, two paths
Start with Observer.
Graduate to Gateway when you need it.
Same registry, same console, same data. Flip Gateway on per workspace when you need it.
Observer
Ships on every tier. Team layer value, no in-path risk.
- Ingress
- Webhooks · OAuth · Artifact reads · OTLP
- Enforcement
- Read-only, advisory
- Setup
- 15-min, no code changes
- Tiers
- Starter · Design Partner · Enterprise
Gateway
In-path enforcement when CISO asks. Same platform, one toggle.
- Ingress
- In-path proxy on model calls
- Enforcement
- Kill switch · Policy · Rate limits
- Setup
- Opt-in per workspace, no migration
- Tiers
- Enterprise only
Gateway is not legacy. Same platform. No migration. No re-onboarding.
Registry
Publish once. Install anywhere.
Skills, agents, playbooks, connectors — versioned, approved, distributed to every teammate's IDE. The team's knowledge compounds here.
$ maiife publish
→ validating manifest…
→ scanning for secrets…
→ submitting to team registry…
✓ triage-incident@1.4.0 approved by @platform-lead
✓ live in 4 workspaces · propagation 2m 41s
→ teammates will load the new version on next session
$ _
Recently in your registry
live- v1.4.0
triage-incident
skill · @mira
- v2.1.3
oncall-routing
agent · @platform
- v3.1.0
incident-response
playbook · @sre
- v1.2.0
salesforce-sync
connector · @revops
- v2.3.1
pr-review
skill · @dev-tools
Security & trust
Privacy-first. Enterprise-ready when you need it.
Defaults already pass review. Deeper governance lives on Enterprise — opt-in, same platform.
Content encrypted at rest
Ask-bar messages, knowledge-base documents and code-translation bodies use AES-256-GCM. Writes fail closed — with no key we refuse to store rather than store in the clear.
Data stays in Singapore
All data is stored in ap-southeast-1. Per-workspace region selection is not offered today.
Audit trail on governance changes
Policy, vendor, kill-switch and approval changes are recorded with before/after state and exportable to CSV.
Individual insights are private by default
A lead sees a member's metrics de-identified until that member opts in, per team.
Work-account sign-in
Google Workspace and GitHub today. Microsoft Entra ID and SAML are in progress for enterprise.
Policy-checked before the model
Every gateway request is authenticated, and proxied completions pass kill-switch, model-access, content-filter, budget, rate-limit and data-routing checks before reaching a vendor.
FAQ
Common questions, straight answers.
What is Maiife?
Maiife is an enterprise AI control plane that lets organizations connect, share, and govern their AI stack. It combines a multi-vendor AI gateway, a policy engine, an artifact registry for team AI knowledge, and team health scorecards in one platform.
What is an AI control plane?
An AI control plane is a governance layer that sits between your teams and AI vendors. It routes requests through a single gateway, enforces policies (model access, spend limits, rate limits, content filtering, data routing, time-based access), and gives leadership visibility into usage, cost, and compliance across every vendor.
How do I govern LLM spend across vendors?
Maiife meters every request across OpenAI, Anthropic, Google, Cohere, and local models, then enforces budgets with alerts at 50%, 75%, 90%, and 100% thresholds. Spend-limit policies can cap teams, users, or models, and kill switches can halt traffic instantly at the global, vendor, team, or model level.
Which AI vendors does Maiife support?
Maiife supports OpenAI, Anthropic, Google, and Cohere in the cloud, plus local model hosting through Ollama and vLLM. Vendors are configured with health checks, circuit breakers, and fallback chains behind one unified API.
How much does Maiife cost?
Maiife is free to use while the platform is in beta — no card required. Pricing has not been finalised yet and will be announced before any plan starts billing. Get in touch if you need enterprise terms in the meantime.
Is there an open-source toolkit?
Yes. Maiife publishes an open-source toolkit under the @maiife-ai-pub npm scope, including probe (vendor health checker), mcp-audit (MCP server auditor), trace, eval, cost, and sub-audit.
Get in Touch
Your AI infrastructure needs a control plane.
Whether you're exploring our open source toolkit or ready for the full platform — we'd love to hear from you.