danm72/mission-control/mission-control-setup
Claude Code and agent skill by danm72/mission-control
- community
Checking that the SKILL.md file is still in the repository.
Install
npx vskill@latest install danm72/mission-control/mission-control-setupSKILL.md
Mission Control Setup Skill
Bootstrap an AI agent squad from Mission Control with a single setup link.
Metadata
| Field | Value | |
|---|---|---|
| name | mission-control-setup | |
| description | Bootstrap an AI agent squad from Mission Control | |
| version | 1.0.0 | |
| author | Mission Control |
Trigger
Activate this skill when the user provides a Mission Control setup URL. Common patterns:
- "Setup my squad using this link: https://missioncontrol.ai/setup/..."
- "Bootstrap agents from https://missioncontrol.ai/api/setup/..."
- "Here's my Mission Control setup link: ..."
- User pastes a URL containing
/setup/or/api/setup/
The setup URL format is:
The setup token should be sent via the x-setup-token header (not as a query parameter).
Environment
No environment variables required. All configuration is fetched from the setup URL.
Paths Used
| Path | Purpose | |
|---|---|---|
~/.openclaw/missions/{squad_id}/ | Squad-specific state and markers | |
~/.openclaw/sessions/{agent_name}/ | Agent session directories | |
~/.openclaw/skills/mission-control/ | Runtime skill installation | |
~/.openclaw/cron/jobs.json | Heartbeat cron job definitions |
Setup Process
Follow these steps in order. Each step is idempotent - safe to re-run.
Step 0: Check for Existing Setup
Before starting, check if this squad was already set up.
If marker exists: Inform the user that setup was previously completed. Show the completion timestamp. Ask if they want to:
- Skip setup entirely
- Re-run setup (will update SOUL.md files and refresh configuration)
If marker does not exist: Proceed with setup.
Step 1: Fetch Configuration
Parse the setup URL and fetch the squad configuration.
Extract URL Parameters
Fetch Configuration
Send the token via the x-setup-token header (preferred over query parameter):
Handle Errors
| HTTP Code | Meaning | Action | |
|---|---|---|---|
| 200 | Success | Proceed to Step 2 | |
| 401 | Invalid token | Tell user the token is invalid. They should generate a new setup link from Mission Control. | |
| 403 | Token expired | Tell user the setup link has expired. They should generate a new one (links expire after 24 hours). | |
| 404 | Squad not found | Tell user the squad ID is invalid or the squad was deleted. | |
| 5xx | Server error | Tell user Mission Control is temporarily unavailable. Try again in a few minutes. | |
| Network error | Connection failed | Check internet connection. Verify the URL is correct. |
On success, parse the JSON response:
Store the config for use in subsequent steps:
Step 2: Create Agent Sessions
Create a session for each agent defined in the configuration.
Using sessions_spawn Tool
For each agent in config.agents:
Idempotency
- Check if session already exists before creating
- If session exists with the same name, skip creation (log that it was skipped)
- New sessions are created in stopped state
Example Output
Step 3: Write SOUL.md Files
Write the SOUL.md file for each agent from the configuration.
Write Files
For each agent in config.agents:
Important Notes
- Always overwrite existing SOUL.md files - this is expected behavior for configuration updates
- The
soul_mdcontent from the API is already complete and ready to write - Create the agent directory if it doesn't exist
Example Output
Step 4: Install Runtime Skill
Install the Mission Control runtime skill that agents will use for heartbeats and task management.
Create Skill Directory
Write Runtime Skill
Write the runtime SKILL.md file that handles ongoing operations:
API_KEY="$MC_API_KEY" AGENT_NAME="$MC_AGENT_NAME"
curl -X POST "https://missioncontrol.ai/api/agents/heartbeat" \ -H "Authorization: Bearer $API_KEY" \ -H "Content-Type: application/json" \ -d "{\"agent\": \"$AGENT_NAME\", \"status\": \"active\"}"
curl -s "https://missioncontrol.ai/api/agents/tasks" \ -H "Authorization: Bearer $MC_API_KEY" \ -H "X-Agent-Name: $MC_AGENT_NAME"
curl -X POST "https://missioncontrol.ai/api/tasks/$TASK_ID/complete" \ -H "Authorization: Bearer $MC_API_KEY" \ -H "Content-Type: application/json" \ -d "{\"agent\": \"$MC_AGENT_NAME\", \"result\": \"$RESULT\"}"
Idempotency
- Overwrite existing skill files - this ensures agents get the latest version
Step 5: Configure Heartbeat Crons
Set up staggered heartbeat cron jobs for each agent.
Calculate Stagger Offsets
Using heartbeat_interval and heartbeat_stagger from config:
Update jobs.json
Job Format
Each cron job should have this structure:
Upsert Logic
- Use composite ID:
{squad_id}-{agent_name}-heartbeat - If job with same ID exists, update it
- If job doesn't exist, add it
- Preserve other jobs in the file
Example Output
Step 6: Verify Setup
Test the API connection for each agent.
Test Each Agent
Handle Verification Failures
If any agent fails verification:
- Report which agents failed and the error
- Common causes:
- Network connectivity issues
- API key was revoked
- Mission Control service is down
- Setup is still considered complete - the issue is likely temporary
- Suggest the user check the Mission Control dashboard
Step 7: Mark Setup Complete
Write the completion marker file.
Completion Report
After successful setup, display a summary:
Error Scenarios
Invalid Setup URL
Expired Token
Network Error
Partial Setup Failure
Re-running Setup
When setup has already been completed and the user wants to re-run:
This is the start of the file. Read the full SKILL.md on GitHub.
Security
- Security scanPassed, 98 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingskills/mission-control-setup/SKILL.md
Popularity
- Repository stars
- 79
- Repository forks
- 14
- 7-day trend score
- 22
Stars and forks are for the whole GitHub repository, not this skill alone. Refreshed 3 Apr 2026, 7 months ago.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/danm72/mission-control/mission-control-setup)