# Harbor > Harbor is an all-in-one agent execution layer. It moves tools, credentials, code execution, memory, artifacts, jobs, apps, workflows, and traces out of the model context into one workspace-scoped runtime across web, API, the hrbr CLI, and the cli-mcp stdio adapter. ## Docs - [Welcome to Harbor](https://docs.tryharbor.ai/documentation/overview): Top-level documentation index covering concepts, agent integration, plugins, guides, and reference. - [Why Harbor](https://docs.tryharbor.ai/documentation/why-harbor): Tools are not enough; agents need a workspace-scoped execution layer. - [Mental Model](https://docs.tryharbor.ai/documentation/mental-model): Plugin then Tool then Run then Trace, defined once. - [Quickstart](https://docs.tryharbor.ai/documentation/quickstart): From signup to first plugin invoke in under five minutes. - [Plugins Overview](https://docs.tryharbor.ai/plugins/overview): 1,000+ integrations across MCP servers, REST/GraphQL APIs, and CLIs. Install in one command, run their tools from any agent. - [Plugin Catalog](https://docs.tryharbor.ai/plugins/catalog): Browse every Harbor plugin. Filter by kind, category, or search. - [Guides Overview](https://docs.tryharbor.ai/guides/overview): Task-oriented walkthroughs. One outcome per page. - [Connect Claude Code](https://docs.tryharbor.ai/guides/get-started/connect-claude-code): Point Claude Desktop at Harbor's MCP server. ## Concepts - [Workspaces](https://docs.tryharbor.ai/documentation/concepts/workspaces): The top-level isolation boundary for every Harbor resource. - [Plugins](https://docs.tryharbor.ai/documentation/concepts/plugins): Connected services Harbor knows how to talk to (MCPs, APIs, CLIs, GraphQL). - [Exec](https://docs.tryharbor.ai/documentation/concepts/exec): One-off TypeScript execution against your workspace tools. - [Functions](https://docs.tryharbor.ai/documentation/concepts/functions): Named, versioned, schedulable TypeScript that runs against your workspace. - [Apps](https://docs.tryharbor.ai/documentation/concepts/apps): Functions deployed at a public URL. The smallest possible web app. - [Skills](https://docs.tryharbor.ai/documentation/concepts/skills): Agent-facing markdown that teaches an LLM when and how to use Harbor. - [Orbit Runtime](https://docs.tryharbor.ai/documentation/concepts/orbit-runtime): Workspace primitives: storage, cache, db, ai, tool discovery, sockets. - [Runs and Traces](https://docs.tryharbor.ai/documentation/concepts/runs-and-traces): Exec paths leave durable, inspectable run and span history. ## For Agents - [For Agents Overview](https://docs.tryharbor.ai/documentation/for-agents/overview): How AI agents consume Harbor docs. - [Connecting Agents](https://docs.tryharbor.ai/documentation/for-agents/connecting): Point Claude Code, Cursor, Codex, Gemini CLI, or any MCP client at Harbor. - [Agent Skills (well-known)](https://docs.tryharbor.ai/documentation/for-agents/agent-skills): Canonical Harbor agent skills served from /.well-known/agent-skills/harbor/. - [Harbor Meta Skill](https://tryharbor.ai/skill.md): The canonical Harbor agent meta-skill body downstream harnesses load on install. - [Agent Skills Index](https://tryharbor.ai/.well-known/agent-skills/index.json): JSON manifest of Harbor agent skills exposed at this domain. - [API Catalog](https://tryharbor.ai/.well-known/api-catalog): RFC 9727 linkset describing Harbor API surfaces. ## Reference - [Glossary](https://docs.tryharbor.ai/documentation/reference/glossary): Every Harbor term, defined once. - [Limits](https://docs.tryharbor.ai/documentation/reference/limits): Plan-level and technical limits across the Harbor surface. - [Security Model](https://docs.tryharbor.ai/documentation/reference/security): Credential vault, sandbox isolation, OAuth trust boundaries. - [MCP Endpoint](https://mcp.tryharbor.ai/mcp): Protected Harbor MCP endpoint with OAuth protected-resource metadata. ## Site - [Homepage](https://tryharbor.ai/): Product overview and signup entry point for Harbor. - [Blog](https://tryharbor.ai/blogs): Engineering and product writing about Harbor and the MCP ecosystem. - [Skills Index](https://tryharbor.ai/skills): Public index of Harbor agent skills exposed at this domain. - [Connect](https://tryharbor.ai/connect): Deep-link surface agent harnesses use to connect users to Harbor. - [Pricing](https://tryharbor.ai/pricing): Harbor pricing tiers — Free, Workspace, Enterprise. Pricing in private beta. - [Enterprise](https://tryharbor.ai/enterprise): SSO via WorkOS / AuthKit, dedicated workspace, per-tool policy governance, audit log export, Workers for Platforms isolation. - [Security](https://tryharbor.ai/security): WorkOS / AuthKit, workspace tenancy, tool policy, KV-backed credential state, audited runs, sandbox isolation. - [Docs](https://tryharbor.ai/docs): Server-side redirect to docs.tryharbor.ai, the canonical Harbor documentation site. - [Templates](https://tryharbor.ai/templates): Starter recipes for hrbr exec — coming soon as GitHub repos. - [Integrations](https://tryharbor.ai/integrations): Browse the MCP plugins Harbor connects to today, sourced from the @hrbr/registry-catalog package. - [Community](https://tryharbor.ai/community): Track Harbor in the open via GitHub, X / Twitter, and the Harbor Discord at https://discord.gg/zMH7XKgaz5. ## Use Cases - [Use Cases Hub](https://tryharbor.ai/use-cases): Workflow-shaped surfaces of the same Harbor execution layer: one workspace, scoped tools, typed exec, artifacts, and traced runs. - [Coding Agent](https://tryharbor.ai/use-cases/coding-agent): Ship coding agents that touch real code, with workspace-scoped tools, traced runs, and auditable actions. - [Ops Agent](https://tryharbor.ai/use-cases/ops-agent): Ops agents with real production reach — coming soon, on the same execution layer as the coding-agent surface. - [Data Agent](https://tryharbor.ai/use-cases/data-agent): Data agents you can actually audit — coming soon, on the same primitives Harbor already exposes to coding agents. - [Customer Ops](https://tryharbor.ai/use-cases/customer-ops): Customer-ops agents, traced end to end — coming soon; the underlying workspace + plugin model is already shipped. ## Compare - [Compare Hub](https://tryharbor.ai/compare): Index of every Harbor comparison page — Composio, Nango, Runlayer, Arcade, Infisical, and Pipedream. - [Harbor vs Composio](https://tryharbor.ai/compare/composio): Comparison of Harbor and Composio on workspace tenancy, OAuth scoping, Cloudflare-native runtime, and GEO bench scores. - [Harbor vs Nango](https://tryharbor.ai/compare/nango): Comparison of Harbor and Nango — agent execution layer vs. integration SDK for backend code. - [Harbor vs Runlayer](https://tryharbor.ai/compare/runlayer): Comparison of Harbor and Runlayer — agent execution layer vs. MCP-security control surface. - [Harbor vs Arcade](https://tryharbor.ai/compare/arcade): Comparison of Harbor and Arcade on workspace tenancy, run/trace audit, and Cloudflare-native runtime. - [Harbor vs Infisical](https://tryharbor.ai/compare/infisical): Comparison of Harbor (agent execution layer) and Infisical (secrets management platform with Agent Vault credential proxy and Agent Sentinel MCP brokering). - [Harbor vs Pipedream](https://tryharbor.ai/compare/pipedream): Comparison of Harbor (all-in-one agent execution layer) and Pipedream (developer-first workflow automation with Pipedream Connect for per-end-user OAuth and hosted MCP servers at mcp.pipedream.com). ## Optional - [Privacy Policy](https://tryharbor.ai/privacy): How Harbor handles user data and workspace content. - [Terms of Service](https://tryharbor.ai/terms): Terms governing use of Harbor.