docs-mcp

Tandem enforces policy between AI agents and the tools, data, memory, and actions they use.

Agents can reason, draft, and propose work. Tandem decides what they are authorized to see, which tools they can call, which actions must pause for approval, what memory/context they can access, and what evidence gets recorded.

This makes Tandem useful when agents touch real company systems: files, repositories, email, MCP tools, customer data, internal docs, production workflows, and long-running automations.

For platform and security teams, Tandem acts as a runtime control plane for agentic systems: scoped tool access, approval gates, permissioned memory, tenant/resource boundaries, and audit evidence.

The model proposes. Tandem enforces.

In governance terms, Tandem manages delegated authority for AI agents at runtime.

What Tandem Does

• Runs AI workflows with durable state instead of transcript-only execution.
• Scopes which built-in tools and MCP connectors are visible at each workflow step.
• Blocks tool calls that fall outside runtime policy before execution.
• Pauses consequential actions for human approval.
• Controls which company memory and context a run can retrieve.
• Records artifacts, tool events, approval decisions, and audit evidence outside the model context window.

Simple Example

An agent may be allowed to draft a customer email, but not send it.

Tandem can expose the draft tool, hide or block the send tool, pause at an approval gate, resume only after a human approves, and record the decision in the audit trail.

What Tandem Is Not

Not this	Instead
Chatbot wrapper	Runtime layer underneath agents and workflows
Agent framework only	Policy layer that controls what agent workflows can see and do
Approval UI only	Runtime enforcement with approvals as one controlled gate
LLM gateway only	Governs workflow state, tools, memory, approvals, artifacts, and audit
Flat RAG system	Runtime-scoped memory and source-bound retrieval
Prompt-only safety layer	Enforcement happens outside the model

Tandem calls this runtime authority: authorization, execution control, approval, scoped memory, and audit enforced outside the model. Entrypoints such as the desktop app, TUI, web control panel, channels, and SDKs are clients of the same engine runtime.

• Runtime-owned controls: Runs, sessions, memory, context, provider secrets, MCP tools, approvals, artifacts, and audit records live outside the model.
• Governed tool execution: Built-in tools and MCP connectors can be scoped per workflow step, with approval gates for consequential actions.
• Tenant-aware runtime: Hosted and enterprise modes carry tenant/principal context through sessions, runs, context runs, memory, provider credentials, MCP secrets, and events.
• Deployable where the data lives: Tandem can run locally, headlessly, hosted, or inside customer infrastructure.
• Provider agnostic: Use OpenRouter, Anthropic, OpenAI, OpenCode Zen, or local Ollama endpoints.

Agent intent -> Runtime policy -> Scoped tool/data access -> Approval gates -> Artifacts -> Audit trail

-> AI runtime infrastructure | Enterprise readiness | Runtime trust boundaries | EU AI Act readiness | Compliance starter pack | Connect an agent via MCP

Why Tandem Exists

Agents are becoming workers. They read company context, call tools, open pull requests, draft customer communication, operate project boards, and prepare decisions that used to stay inside human-only systems.

Prompts are not permissions. A system prompt can ask a model to avoid a tool, skip a folder, or wait for approval, but the model should not be the security boundary. Tandem puts those controls in the runtime, so a workflow can grant the agent only the tools, memory, and actions needed for the current step — and deny anything outside that scope.

Companies also need central AI context without flat access. A permissioned company memory should know what the company knows, but an agent acting for one team, tenant, project, or user should only retrieve the slice it is allowed to use.

What Tandem Governs

• Company knowledge and memory: Runtime-owned memory, knowledge spaces, and retrieval paths designed around tenant and workspace boundaries.
• Tool and MCP visibility: Step-scoped built-in tools and MCP connector tools, with broader pre-invocation masking planned for enterprise deployments.
• Workflow execution: Durable automation and context-run state instead of fragile transcript-only execution.
• Human approvals: Runtime gates pause runs, collect approve/rework/cancel decisions, and leave evidence.
• Tenant and workspace boundaries: Tenant-aware sessions, runs, context runs, events, provider credentials, MCP secrets, memory, and contract vocabulary for resource scopes and grants.
• Connector credentials and secrets: Provider and MCP secret references are runtime-owned; connector source binding gives scoped ingestion a shared contract as that layer matures.
• Artifacts and audit trails: Outputs, validations, tool ledger events, approval decisions, and protected audit records survive outside the model context window.

Core Use Cases

Use case	What Tandem adds
Approval-gated email and workflows	Agent proposes work, Tandem pauses before the action, a human approves or requests rework.
Permissioned company knowledgebase	Company memory and knowledge spaces with tenant-aware retrieval and resource-grant vocabulary.
Governed coding agents	Coder runs, worktree context, handoff artifacts, approval points, and auditable implementation state.
Project, sprint, and event brain	Long-running context, tasks, artifacts, and memory that survive across sessions and teams.
Tenant-isolated hosted automations	Hosted runtime records, event streams, provider credentials, MCP secrets, and memory scoped by tenant.
Internal agent and tool governance	A control point for which agents can see which tools, execute which actions, and leave which evidence.

Why Platform And Security Teams Care

Tandem is designed for teams that need to run AI work under real operational controls:

• Runtime authority, not prompt authority: The model can request context or a tool call; the runtime decides what is visible and executable.
• Tenant-aware records: Sessions, automation runs, context runs, event streams, provider credentials, MCP secrets, and memory paths carry tenant context in hosted/shared modes.
• Resource and grant model: Tandem models resources, principals, grants, data classes, and data boundaries so access decisions can be enforced by the runtime.
• Permissioned memory: Memory and knowledge paths carry tenant boundaries so company knowledge can become useful without becoming globally flat.
• Deployable runtime: The same runtime can run on a laptop, as a headless engine, hosted, or inside customer infrastructure as the enterprise layer matures.
• Auditability: Approval decisions, policy denials, provider secret changes, MCP activity, tool ledger events, artifacts, and protected audit records can be inspected outside chat transcripts.

Deployment Model

Tandem is useful locally and grows toward stricter company deployments:

• Local desktop: Single-user desktop runtime with local workspace scope, provider setup, and approval-gated tools.
• Headless engine: tandem-engine serve for SDKs, control panels, automations, and CI/dev environments.
• Hosted/private managed: Hosted deployments with transport-token and signed context assertions for tenant-aware access.
• Customer infrastructure: A deployment model for running where company data, connector credentials, and operational evidence need to live.

Current Status

Current capabilities	Enterprise roadmap
Runtime auth modes: local_single_tenant, hosted_single_tenant, enterprise_required	Full RBAC, OIDC, SCIM, SIEM integrations, SOC2 package, and enterprise identity policy bridge
Tenant context and signed context assertions for hosted/enterprise ingress	Private enterprise sidecar with fail-closed policy authorization
Tenant-aware sessions, automation runs, context runs, events, coder routes, and memory APIs	Complete artifact/export isolation across every path
Provider credential and MCP secret tenant boundaries	Full tool-discovery masking before model invocation
Memory tenant partitioning, tenant-scoped knowledge spaces, and resource-scoped retrieval APIs	Production connector ingestion admin platform with live external source ingestion
Resource access-control contract types and strict context projection vocabulary	Signed approval receipts and auditor-grade immutable receipt chains
Approval gates, pending approval inbox, channel approvals, tool ledger events, and audit records	Advanced connector quarantine/revoke/rotate operations wired to production ingestion workflows

Compliance and AI Act readiness

Tandem helps teams operate AI workflows with human oversight, scoped tools, durable execution evidence, and protected-action controls. For regulated or security-sensitive deployments, start with the EU AI Act readiness brief, then use the Compliance Starter Pack for control mapping, Article 50 transparency guidance, deployer instructions, an Annex IV documentation template, and a limitations/responsibility matrix.

30-second quickstart

Web Control Panel

Install the master CLI, then bootstrap the panel and its engine service:

npm i -g @frumu/tandem
tandem install panel
tandem panel init
tandem panel open

Use this when you want the browser-based control center backed by the engine.

For local installs, you can now open Settings -> Providers -> openai-codex and choose Connect Codex Account to sign in through the browser instead of pasting an OpenAI API key.

Desktop

1. Download and launch Tandem: tandem.ac
2. Open Settings and add a provider API key, or use the local control panel to connect a Codex account for openai-codex.
3. Select a workspace folder.
4. Start with a task prompt and choose Immediate or Plan Mode.

Editable App Scaffold

Generate a fully editable control panel app in your own folder:

npm create tandem-panel@latest my-panel
cd my-panel
npm install
npm run dev

Use this when you want to customize routes, pages, themes, styles, or runtime behavior without editing node_modules.

MCP-assisted setup

If you want an existing agent to help install or configure Tandem, connect that agent to Tandem's MCP interface first. The MCP docs explain how to wire your own agent into Tandem so it can assist with setup, configuration, and follow-up tasks:

• Tandem MCP docs

If you only want the engine runtime, you can keep it foreground-only:

tandem-engine serve --hostname 127.0.0.1 --port 39731

Other Entry Points

• TUI: npm i -g @frumu/tandem-tui && tandem-tui
• SDKs: npm install @frumu/tandem-client or pip install tandem-client

Codex And Docker Setup

• Codex users can connect Tandem through the tandem-codex-plugin repository.
• For a Docker-based Tandem agents setup, see tandem-agents.

Open Core & Source-Available Architecture

Tandem is built for developers first, using an open-core model. We believe that to trust an AI runtime, you must be able to audit the execution router line-by-line.

Local Development & Evaluation: The permissively licensed crates and libraries (MIT OR Apache-2.0) may be used under their own terms. Every distributed engine binary also includes the source-available BUSL-1.1 components, which are free for evaluation, development, testing, source inspection, personal non-commercial use, and non-production proofs of concept.

Enterprise Path: Advanced features for scaled organizational deployments, such as enterprise identity federation, richer policy enforcement, signed receipt chains, private sidecar enforcement, SIEM export, and HA packaging, are planned enterprise capabilities and may be governed under commercial or source-available terms, including the Business Source License 1.1 (BUSL-1.1) where declared.

License Boundary: Commercial production use of the BUSL-1.1 components — including internal production use, client production deployments, and managed, hosted, SaaS, white-label, embedded, OEM, or reseller offerings — requires a separate commercial license from Frumu LTD. See docs/LICENSING.md for the exact package-by-package terms.

Architecture

flowchart TD
    Human[Human operator or team]

    subgraph Entrypoints["Entrypoints: clients, not authority boundaries"]
        Desktop[Desktop app]
        Panel[Web control panel]
        TUI[Terminal UI]
        SDK[TypeScript / Python SDKs]
        Channels[Slack / Discord / Telegram]
    end

    subgraph Agents["Agents and models: propose, reason, draft"]
        Workers[Agent workers]
        Models[OpenAI / Anthropic / OpenRouter<br/>OpenCode Zen / Ollama]
    end

    subgraph Tandem["Tandem governed runtime: owns authority"]
        API[HTTP/SSE API]
        Tenant[Auth mode, tenant context<br/>and authority chain]
        Projection[Authority projection<br/>resources, grants, data classes]
        Runs[Sessions, workflows<br/>automations, context runs]
        Gates[Human approval gates]
        Policy[Tool and MCP policy]
        Memory[Permissioned memory<br/>and company knowledge]
        Secrets[Provider and MCP secrets]
        Artifacts[Artifacts, validation<br/>and run evidence]
        Audit[Audit trail and tool ledger]
    end

    subgraph Systems["Company systems and data"]
        Workspace[Workspace files and repos]
        MCP[MCP servers and connectors]
        Data[Customer / company data]
        Browser[Browser and external tools]
    end

    Human --> Desktop
    Human --> Panel
    Human --> TUI
    Human --> SDK
    Human --> Channels

    Desktop --> API
    Panel --> API
    TUI --> API
    SDK --> API
    Channels --> API

    API --> Tenant
    Tenant --> Projection
    Projection --> Runs
    Runs <--> Workers
    Workers <--> Models

    Runs --> Gates
    Runs --> Policy
    Runs --> Memory
    Runs --> Artifacts
    Gates --> Audit
    Policy --> Audit
    Artifacts --> Audit

    Policy --> Secrets
    Policy --> Workspace
    Policy --> MCP
    Policy --> Data
    Policy --> Browser
    Memory --> Data

Common workflows

Governed workflow	What the Tandem runtime controls
Evaluate vendor or policy risk	Read selected sources, draft cited artifacts, validate limitations, and keep mutation tools outside the read step.
Approval-gated email or updates	Let an agent draft the action, pause at a human gate, resume only after approve/rework/cancel evidence is recorded.
Execute code migrations	Track coder runs, worktree state, changed files, validation, handoff artifacts, and approval points.
Govern external MCP tools	Scope connector tools by workflow step, require concrete tool evidence, and isolate MCP secrets by tenant path.
Permissioned company memory	Retrieve company context through runtime-owned memory and knowledge spaces instead of pasting everything into chat.
Tenant-isolated hosted workflows	Keep sessions, runs, events, credentials, MCP secrets, and memory partitioned by tenant in hosted/shared modes.

Features

Governed execution

• Model is not the control system: The model can propose work; Tandem owns the allowed tools, context, state transitions, approvals, and audit evidence.
• Scoped workflow execution: Automation V2 nodes can carry built-in tool and MCP connector policy so different steps see different capabilities.
• Approval-gated actions: Runs can halt before consequential work, wait for approve/rework/cancel decisions, and resume with recorded gate history.
• State Survival: Checkpoints, replayable event history, and materialized run states that survive API timeouts and connector failures.

Permissioned memory and company knowledge

• Tenant-partitioned memory: Vector-backed session, project, global, and file-import memory paths carry tenant scope.
• Knowledge spaces: Curated knowledge spaces and items can be managed through tenant-aware APIs.
• Runtime retrieval: Agents retrieve context through the runtime, creating a path toward permissioned company memory instead of flat transcript stuffing.

Tenant and workspace isolation

• Tenant-aware records: Sessions, automation definitions/runs, context runs, event streams, provider credentials, MCP secret references, memory, and coder routes carry tenant context.
• Strict context contract: Resource refs, scoped grants, data classes, principals, and data boundaries are modeled in the enterprise contract.
• Local behavior preserved: Local/default single-tenant usage remains the default for desktop and developer workflows.

MCP and tool governance

• MCP connector support: Tandem can connect to MCP servers, sync tools, and scope selected connector tools to workflow nodes.
• Secret isolation: Store-backed MCP secret references validate tenant scope before resolution in hosted/shared paths.
• Tool discovery as an authority surface: Tool discovery and MCP visibility are governed by runtime policy; full masking of unauthorized tools before model invocation remains enterprise roadmap work.

Human approval gates and audit

• Approvals inbox and channel approvals: Operators can approve, request rework, or cancel from runtime-owned approval surfaces.
• Durable evidence: Tool ledger events, gate history, artifacts, validation metadata, and protected audit events can survive outside the model context window.
• Audit stream: Admin-facing audit streams and protected audit envelopes exist; immutable receipt chains and signed approval receipts remain roadmap work.

Multi-agent orchestration

• Kanban-driven execution: Agents claim tasks, report blockers, and hand off work through deterministic state.
• Memory-aware workers: Agents can use prior run context, project memory, and knowledge spaces through runtime paths.
• Revisioned coordination: Engine-enforced locks prevent agents from trampling the same codebase simultaneously.

Local and self-hosted controls

• MCP tool connectors
• Scheduled automations and routines
• Headless runtime with HTTP + SSE APIs
• Desktop runtime for Windows, macOS, and Linux
• API keys encrypted in local SecureKeyStore (AES-256-GCM)
• Local Codex OAuth credentials stay engine-owned; browser UIs initiate sign-in but do not persist refresh tokens
• Workspace access is scoped to folders you explicitly grant
• Write/delete operations require approval via supervised tool flow
• Sensitive paths denied by default (.env, .ssh/*, *.pem, *.key, secrets folders)
• No analytics or call-home telemetry from Tandem itself

Outputs and artifacts

• Markdown reports
• HTML dashboards
• PowerPoint (.pptx) generation

Enterprise Path And Roadmap

Tandem already includes the runtime building blocks for governed AI work in hosted and self-managed environments. The next enterprise capabilities strengthen identity, policy, audit export, and administration around those building blocks.

Available now:

• Runtime auth modes and hosted/enterprise signed context assertion verification.
• Tenant-aware sessions, runs, context runs, event streams, provider credentials, MCP secrets, memory, and coder routes.
• Resource access-control contract vocabulary for resources, scopes, principals, grants, data classes, and data boundaries.
• Approval gates, protected audit records, audit streams, tool ledger events, and runtime artifacts.
• Connector source-binding contracts using secret references, resource refs, data classes, quarantine/revoke/rotate vocabulary, and scoped memory chunk references.

Planned enterprise capabilities: