# Terse - [Quickstart](https://docs.useterse.ai/quickstart.md): Create, test, and deploy your first Terse project - [Context as Code](https://docs.useterse.ai/context-as-code.md): Terse compiles your workspace into a typed SDK. Workflows reference real resources by name, and coding agents build against ground truth instead of guesses. - [Hosting overview](https://docs.useterse.ai/hosting.md): Where Terse runs: the control plane (orchestrator, triggers, UI, history) and the data plane (where your job code executes). Each can run on Terse Cloud or in your own infrastructure. - [Self-hosting the data plane](https://docs.useterse.ai/self-hosting.md): Run your workflow handlers in your own infrastructure while Terse Cloud continues to handle triggers, scheduling, and run history (Hybrid deployment). - [Self-hosting the control plane](https://docs.useterse.ai/self-hosting-control-plane.md): Run the Terse orchestrator inside your own network with one command. - [Projects](https://docs.useterse.ai/core-concepts/projects.md): A project is the deployable unit in Terse: a TypeScript codebase with one or more jobs that ships as a single bundle. - [Jobs](https://docs.useterse.ai/core-concepts/jobs.md): A job is the unit of automation in Terse: a trigger and a handler that runs every time the trigger fires. - [Skills](https://docs.useterse.ai/core-concepts/skills-and-integrations.md): Connect external services, generate typed code, and control what the model can reach. - [Generated SDK](https://docs.useterse.ai/core-concepts/generated-sdk.md): Anatomy of src/terse.generated.ts and the terse.generated/ folder. What terse generate writes, how your workflows import it, and why it's the single biggest ergonomic win for both humans and coding agents. - [TerseAgent](https://docs.useterse.ai/core-concepts/terse-agent.md): Runs agentic loops, returns structured outputs, and enforces a strict allowlist of skills and resources. - [Deterministic tool calls](https://docs.useterse.ai/core-concepts/deterministic-tool-calls.md): Call integration tools directly from your workflow code with guaranteed execution, no LLM in the loop. - [Human-in-the-loop](https://docs.useterse.ai/core-concepts/human-in-the-loop.md): Require human approval before specific tools execute, whether in the terminal during development or via Slack, email, and the app in production. - [Durability](https://docs.useterse.ai/core-concepts/durability.md): Opt-in durable execution: jobs that pause, wait for as long as you need, and survive restarts without repeating work. - [Activity & observability](https://docs.useterse.ai/core-concepts/activity-and-observability.md): Every workflow run is recorded with a full trace of actions taken, model decisions, failures, and timing. - [Self-improvement](https://docs.useterse.ai/core-concepts/self-improvement.md): Terse reviews your runs every week and ships patches against your workflow code. - [CLI reference](https://docs.useterse.ai/reference/cli.md): Reference for the terse CLI - [Triggers](https://docs.useterse.ai/reference/triggers.md): All available triggers that can start a Terse workflow. - [Skills & Tools](https://docs.useterse.ai/reference/skills.md): All available skills and the tools they expose to your workflows. - [TypeScript SDK reference](https://docs.useterse.ai/reference/typescript-sdk.md): Reference for the terse-sdk package and the generated helpers used in Terse workflows. - [AGENTS](https://docs.useterse.ai/AGENTS.md) - [Compare](https://docs.useterse.ai/compare/index.md): See when Terse fits better than Zapier, raw scripts, or fully autonomous agent platforms. - [Overview](https://docs.useterse.ai/index.md): Terse is the AI workflow platform for coding agents. - [Reference](https://docs.useterse.ai/reference/index.md) - [Lead generation pipeline](https://docs.useterse.ai/use-cases/lead-gen-pipeline.md): Enrich, score, and route new leads into Attio on autopilot.