Continue Mission Control - Continue

What You Can Do in Mission Control

Mission Control is where developers and teams use Continue to power everyday coding workflows. From quick one-off Tasks to fully automated pipelines, everything starts here with Cloud Agents (AI-powered).

<CardGroup cols={3}> <Card title="Beyond Checks" icon="list" href="/mission-control/beyond-checks"> Agents, triggers, and automation beyond PR reviews. </Card> <Card title="Manage Your Cloud Agents" icon="robot" href="/mission-control/beyond-checks"> Configure, run, and monitor your custom Cloud Agents across repositories. </Card> <Card title="Connect Integrations" icon="plug" href="/mission-control/integrations"> Link GitHub, Slack, Sentry, and more to power your Cloud Agents. </Card> </CardGroup>

Inbox: One Place to Triage, Assign, and Take Action

<Info> The Inbox is your team's central queue for all Cloud Agent powered Tasks, activity, and first-class integrations like Sentry and Snyk. It gives developers a shared place to review work, delegate responsibility, and launch follow-up Tasks—all without leaving Mission Control. </Info> <CardGroup cols={4}> <Card title="Unified View" icon="list"> See every incoming Task, Sentry alert, Snyk vulnerability, and workflow output in one place. </Card> <Card title="Powerful Filtering" icon="filter"> Slice the Inbox by Agent, repository, integration, creator, status, or time to surface what matters now. </Card> <Card title="Clear Ownership" icon="user-check"> Assign Sessions to teammates so responsibilities are explicit and reviews move forward. </Card> <Card title="Take Action Fast" icon="zap"> Launch follow-up Tasks, investigate errors, or generate fixes directly from the Inbox—no context switching. </Card> </CardGroup>

First-Class Integration Tabs

Some signals deserve their own dedicated view. When you connect integrations:

<AccordionGroup> <Accordion title="Sentry"> See error events organized by project with occurrences, first/last seen timestamps, and one-click **Solve** actions that launch an Agent to investigate or generate a fix. </Accordion> <Accordion title="Snyk"> Review high-severity vulnerabilities, check CVSS scores, and trigger Agents to patch or upgrade dependencies. </Accordion> </AccordionGroup>

Monitoring & Insights

<Info> Track performance, share configurations across your organization, and monitor Cloud Agent activity in real time. </Info> <CardGroup cols={3}> <Card title="Metrics" icon="chart-column" href="/mission-control/metrics"> Measure success rates, PR activity, and intervention trends across Cloud Agents. </Card> <Card title="Organization Sharing" icon="users" href="/mission-control/governance/org-permissions"> Share Cloud Agents with teammates using built-in access controls. </Card> <Card title="Cloud Agent Monitoring" icon="radar" href="/mission-control/beyond-checks#agents"> View active runs, logs, and outcomes for every Cloud Agent session. </Card> </CardGroup> ---

Components Behind Mission Control's Cloud Agents

Cloud Agents are built from reusable components that you can create, share, and customize:

Large Language Models from various providers (OpenAI, Anthropic, etc.) configured for specific roles like chat, autocomplete, or editing

</Card> <Card title="Tools" icon="wrench">

Model Context Protocol servers that provide tools and capabilities like database access, web search, or custom functions

</Card> <Card title="Rules" icon="shield">

Guidelines that shape AI Agent behavior - coding standards, constraints, or specific instructions for your domain

</Card> <Card title="Prompts" icon="message-square">

Reusable instructions for common tasks, optimized for specific workflows or coding patterns

</Card> </CardGroup> <Tip>

When you create a component for your Cloud Agent in Mission Control, it becomes available according to the permissions you set: Personal, Public, or Organization.

</Tip> <Accordion title="Component Inputs">

Some components can receive values, including secrets, as inputs through templating. For values that the user needs to set, you can use template variables (e.g. ${{ inputs.API_KEY}}). Then, the user can set API_KEY: ${{ secrets.MY_API_KEY }} in the with clause of their agent or config.

<Info> **Choosing between `secrets.` and `inputs.`**

When creating components for your Agent:

Use ${{ inputs.INPUT_NAME }} in your definition when you want users to be able to customize which secret is used
Users will then map their own secrets using ${{ secrets.SECRET_NAME }} in the with clause

For personal or single-use configurations, you can skip the inputs layer and reference ${{ secrets.SECRET_NAME }} directly. </Info> </Accordion>

Get Started

<Frame> <CardGroup cols={3}> <Card title="Create a Task" icon="plus" href="https://continue.dev"> Start with a prompt and run an Agent on demand. </Card> <Card title="Build Workflows for Cloud Agents" icon="clock" href="https://continue.dev/agents/new"> Automate work using GitHub triggers, cron schedules, or webhooks. </Card> <Card title="Add Integrations" icon="workflow" href="https://continue.dev/integrations"> Connect Slack, Sentry, Snyk and more to extend your Agents. </Card> </CardGroup> </Frame>