DocsDeveloper quickstart

Start your agentic team in five steps.

The SICKR CLI runs configured agents on your machine or runner. Identity, repository policy, workflows, approvals, evidence, live operations, audit, insights, and billing are managed in Sickr Terminal.

Prerequisite: bring your own agent. You bring the agents. SICKR drives them on your own Claude Code, Codex, Gemini, or local runtime. It is the governance and orchestration layer, not a model reseller.

  1. Step 01

    Start — Claim your organization

    Sign in with GitHub as the invited owner, then connect the GitHub organization the work runs in.

    Your engagement is agreed before you start, and includes the complete product. Agent capacity and workflow scope follow that agreement rather than a plan you pick here.

  2. Step 02

    Install — Prepare the local runtime

    Use one SICKR CLI to connect configured agents to Sickr Terminal for governed execution, live monitoring, intervention, and audit.

    Install git and GitHub CLI, authenticate gh, and install the agent runtime you want SICKR to drive: Claude Code, Codex, Gemini CLI, or an approved local runtime. The first start checks these prerequisites and the required SICKR MCP connection.

    install
    # no global SICKR install required
gh auth status

  3. Step 03

    Configure — Agents in Sickr Terminal

    Create and configure each agent in workflow.sickr.ai/admin/agents. The registered agent id is the id you pass to the local CLI.

    Choose the agent CLI, model, repositories, skills, and governed capabilities for each agent. SICKR owns the local process so it can dispatch work, stream activity, enforce controls, and retain an accountable execution record.

    agents
    # Configure agents on workflow.sickr.ai/admin/agents
# Example ids:
#   codex-001  -> Codex-backed agent
#   claude-001 -> Claude-backed agent

  4. Step 04

    Kindle — Start agents

    Start one local background process per configured agent. Each process connects using the agent id registered in Sickr Terminal.

    The command frees your terminal, fetches a short-lived runtime lease, starts the configured agent, and keeps it ready for governed work. Live monitoring, replay, intervention, and workflow execution are capabilities of the same SICKR product.

    kindle
    npx @sickr/cli start --agent-id codex-001
npx @sickr/cli start --agent-id claude-001

# stop one or all background agents
npx @sickr/cli stop --agent-id codex-001
npx @sickr/cli stop all

  5. Step 05

    Run — Work through Sickr Terminal

    Create tickets in Sickr Terminal and watch configured agents move work from intake to a governed outcome.

    SICKR plans, reviews, implements, validates, records, and advances each ticket through the workflow you choose. Humans can approve, supervise, intervene, or recover work in the same accountable flow.

Troubleshooting

  • 401 / unauthorized — confirm the agent id exists on workflow.sickr.ai/admin/agents and that this local machine is connected for that agent.
  • agent won't start — check that git, gh, and the selected agent runtime are on PATH, then retry npx @sickr/cli start --agent-id <agent-id>.
  • MCP initialization failed — run the selected agent CLI directly once, complete its SICKR MCP setup, and confirm the SICKR server initializes before restarting the agent.
  • capacity exceeded — the organization member has reached the active-agent capacity for the current plan. Stop an unused agent or change plans in Terminal billing.
Talk to us →