← Back to Skills Marketplace
130
Downloads
0
Stars
0
Active Installs
5
Versions
Install in OpenClaw
/install tsu-mission-control
Description
Integrate with Mission Control dashboard to report task progress, publish documents to the Library, request approvals, and submit project requests. Use this...
Usage Guidance
This skill appears to be what it advertises: a self-hosted dashboard plus an OpenClaw lifecycle hook. Before installing, consider: 1) The setup runs npm install and may pull many third-party Node packages — review package.json/dependencies if you need to limit supply-chain risk. 2) The hook will forward OpenClaw lifecycle events (session keys, tool calls, and any event payloads) to the configured MISSION_CONTROL_URL. If agents include file contents or sensitive session/cost data in events, those will be sent to the dashboard — set a strong HOOK_SECRET and review what agents are allowed to send. 3) The installer and docs provide ways to bind the frontend/backend to the LAN and to run systemd units; only enable those if you intend network exposure and have configured CORS and firewall rules. 4) If you’re uncertain, run the dashboard in Docker or in an isolated environment, inspect installed npm packages, and do not enable the hook in OpenClaw until you’re ready. If you want, I can list the top package.json dependencies to review or point out the exact places agents could unintentionally send local files or credentials.
Capability Analysis
Type: OpenClaw Skill
Name: tsu-mission-control
Version: 1.1.2
The skill bundle implements a comprehensive management dashboard that includes a 'Dispatcher' service (app/backend/src/services/dispatcher.js) capable of sending natural-language instructions and 'ACTION REQUIRED' prompts back to the AI agent. While aligned with the stated purpose of workflow orchestration, this creates a remote influence vector over the agent's behavior. Additionally, the backend uses custom regex-based HTML sanitization (app/backend/src/middleware.js), which is a vulnerable pattern for preventing XSS. The default unauthenticated configuration and the ability to remotely trigger agent actions via the OpenClaw gateway represent risky capabilities without clear evidence of malice.
Capability Assessment
Purpose & Capability
Name/description (dashboard, task reporting, library, approvals) match the included files: a backend (Express + SQLite), frontend, lifecycle hook, setup script and full docs. Required binaries (node, npm) are appropriate. The metadata/installation reference to a Node package (mission-control-skill) is consistent with a Node-based install even though full source is included.
Instruction Scope
SKILL.md documents HTTP endpoints and example event payloads (task lifecycle, library publish, approvals). That is within scope. Important notes: examples include file paths and content fields (e.g. source_path: /workspace/docs/... and library:publish content payloads), and session events may include token/cost data in payloads; the lifecycle hook forwards whatever event/data OpenClaw supplies to the configured MISSION_CONTROL_URL. That means agents could send (and the hook will transmit) workspace file contents or session cost/token details if included in events — expected for this integration but a potential data-exfiltration vector if misused.
Install Mechanism
Installation is local: included setup.sh copies the provided app into ~/.openclaw/mission-control, generates a secret and creates backend .env, or can use Docker Compose. The package will run npm to install dependencies (normal for Node apps). There are no downloads from arbitrary shorteners or personal IPs in the provided files. Note: npm install will fetch packages from the public registry — review dependencies if you require strict supply-chain control.
Credentials
The skill itself does not demand unrelated cloud credentials. The install and runtime expect/produce integration secrets: MISSION_CONTROL_HOOK_SECRET and an optional OPENCLAW_GATEWAY_TOKEN (stored in the dashboard .env) so the backend can authenticate with the gateway and validate hook posts. Those are proportional to a dashboard↔agent integration. Be aware that the hook will forward session and tool-call data (including the sessionKey and whatever event payloads OpenClaw provides), so any sensitive data present in events or agent-provided deliverables could be sent to the dashboard backend.
Persistence & Privilege
The skill installs files under the user's OpenClaw directory and copies a hook into ~/.openclaw/hooks and suggests adding an entry to openclaw.json and optionally enabling systemd services or binding to LAN. always:false (no forced inclusion) and autonomous invocation enabled (default) — expected for an integration hook. Because it adds a lifecycle hook and may run a local server, this integration has deeper system presence than instruction-only skills; that's normal but raises the usual operational exposure (service listening ports, CORS, systemd units) which you should control.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install tsu-mission-control - After installation, invoke the skill by name or use
/tsu-mission-control - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.1.2
- Added database migration script: 005_pending_dispatches.sql.
- Introduced new dispatch route handler at app/backend/src/routes/dispatch.js.
- Foundation laid for polling or handling "pending dispatches" (see final section in documentation).
- No changes to existing API contract or endpoints.
v1.1.1
- Changed the authentication header for API requests from Authorization: Bearer to X-Hook-Secret.
- No other content or functionality changes detected.
v1.0.2
Version 1.0.2
- Added backend middleware and dispatcher service for event handling and workflow integration.
- Introduced frontend utilities: usePagination hook, MarkdownRenderer and icons components, and notifications handler.
- Added a dispatcher test file for backend testing purposes.
- Added a changelog document for tracking changes.
- Improved session lifecycle documentation in the SKILL.md, including session:created event usage.
v1.0.1
Initial public release of Mission Control skill with full integration backend and frontend.
- Added backend implementation for Mission Control APIs, including agents, tasks, approvals, reviews, projects, requests, events, costs, and library management.
- Added frontend scaffolding and configuration for dashboard UI.
- Implemented complete database schema migrations for core features.
- Included documentation detailing API endpoints, event formats, and skill usage best practices.
- Provided Docker support and environment example templates for local deployment and testing.
v1.0.0
Initial release: Mission Control integration skill.
- Allows agents to report task progress, completion, failure, and request review through standardized HTTP events.
- Enables publishing and updating documents to an organized Library with support for research, reports, documentation, and more.
- Supports requesting human approvals with workflow gates and resume tokens.
- Lets agents submit new project requests (features, bugs, research, etc.).
- Reports session costs (token usage) and agent status (idle, errors).
- Provides /mc shorthand commands for simplified reporting in supported environments.
Metadata
Frequently Asked Questions
What is Mission Control?
Integrate with Mission Control dashboard to report task progress, publish documents to the Library, request approvals, and submit project requests. Use this... It is an AI Agent Skill for Claude Code / OpenClaw, with 130 downloads so far.
How do I install Mission Control?
Run "/install tsu-mission-control" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Mission Control free?
Yes, Mission Control is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Mission Control support?
Mission Control is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Mission Control?
It is built and maintained by louisval1 (@louisval1); the current version is v1.1.2.
More Skills