AI + Operately

Your AI agent can use Operately like you do

Operately has a full CLI, API, and skills that let AI agents create goals, update projects, and post check-ins. If your agent can run shell commands, it can operate Operately the same way a human would.

Full API CLI for agents Published skills Open source

How AI agents work with Operately

Operately exposes every operation through a CLI and API. When an AI agent like OpenClaw, Codex, or Claude Code needs to interact with your goals and projects, it authenticates with a token and runs the same commands a human would in the terminal.

There is no special "AI mode." An agent that can run shell commands is already a full Operately user. It can create goals, update project status, write check-ins, and query progress. The structured data (typed targets, milestones, status fields) makes it easy for agents to read and reason about your team's work.

The toolchain

CLI

Install the Operately CLI (@operately/operately-cli) from npm. Authenticate with a token. Every operation is a command: goals, projects, check-ins, people, spaces.

Agent skills

Operately publishes agent skills that teach compatible tools how to work with Operately workflows. Install with npx skills add and your agent already knows the commands. Also available on ClawHub for OpenClaw.

API

The full external API is documented and stable. Everything the web interface can do, the API can do. Agents can query goals, post check-ins, and update project status programmatically.

Structured data

Goals have measurable targets. Projects have milestones and owners. Check-ins have status fields. Everything is structured, not freeform text. Agents can read, reason about, and update this data without parsing documents.

What this looks like in practice

# Install the CLI
npm install -g @operately/operately-cli

# Authenticate
operately auth login --token $OPERATELY_TOKEN

# An agent creates a goal
operately goals create --name "Ship v2 by March" --target "100%" --space engineering

# An agent posts a weekly check-in
operately projects check_in --project website-redesign --status on_track \
  --description "Completed homepage, starting landing pages next week"

# An agent reviews goal progress
operately goals list --space engineering

Any agent that can run shell commands can operate Operately. No custom integrations, no proprietary SDKs, no API wrappers to maintain.

Works with the agents you already use

Operately integrates with AI agents that can execute commands and interact with APIs:

  • OpenClaw — Install the Operately skill and your OpenClaw agent manages goals and projects alongside everything else it does for your team.
  • OpenAI Codex — Codex can use the CLI to update project progress after completing coding tasks.
  • Claude Code — Same approach: install skills, authenticate, and Claude Code becomes a full Operately participant.
  • Any tool with shell access — If it can run operately commands, it works.

Agent-readiness compared

Operately Notion Monday / Asana
CLI Yes No No
Official agent/MCP tooling Yes (skills) Yes (MCP server) No
API Yes Yes Yes
Open source Yes No No
Self-hostable Yes No No
Built-in goals/OKRs Yes No Paid add-on

Why this approach works

Most tools add AI as a feature inside their interface: a copilot that summarizes, a chatbot that answers questions about your data. That helps, but it is limited to what the vendor decides to build.

Operately takes a different approach. The CLI and API treat agents the same way they treat humans. An agent authenticated with a token has the same capabilities as someone logged into the web interface. This means any new agent tool, any new AI framework, any custom automation you build can immediately work with Operately without waiting for us to build an integration.

Your coding agent finishes a milestone? It updates the project. Your operations agent does a weekly review? It posts a check-in. The tool gets out of the way and lets agents participate in your team's workflow naturally.

Try Operately with your agents

Free to start. Full API access on every plan. Open source, self-hostable, ready in minutes.

Start free on Cloud Explore the CLI docs