A persistent, cloud-based storage solution for AI agent memory, logs, and secrets using a managed Convex backend.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install launchthatbot-convex-backend
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install launchthatbot-convex-backend using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Convex Backend?
The Convex Backend skill provides a robust alternative to local file storage by integrating your AI agents directly with Convex. This skill allows for durable long-term memory, structured daily logs, and secure environment variable management that survives container restarts and environment migrations. By utilizing Openclaw Skills like this one, developers can build agents that maintain context across sessions and share data safely in multi-agent architectures.
This integration uses the stock Convex MCP server, meaning you maintain full ownership of your data and infrastructure while benefiting from a managed control plane. It is particularly valuable for users of Openclaw Skills who require high availability and professional-grade security for their agentic workflows.
Convex Backend Use Cases
- Persisting agent memory and user preferences across system restarts.
- Centralizing secrets and API keys in a secure cloud environment instead of local .env files.
- Sharing context and logs between multiple agents working on a single project.
- Automating the generation of structured daily logs for project tracking and visibility.
How Convex Backend Works
- The skill connects to your project using a CONVEX_DEPLOY_KEY provided in the local environment.
- It leverages the stock Convex MCP server to execute functions on your backend.
- Upon initialization, it can migrate existing local secrets into the Convex environment for better security.
- During operation, the agent uses the memory:addMemory function to store facts, preferences, and decisions.
- At the end of each session, the agent writes a summary to the daily log to ensure continuity.
- This ecosystem of Openclaw Skills allows for seamless retrieval of past context via searchMemory and getDailyLog functions.
Convex Backend Setup
To get started with this skill, ensure you have a Convex account and a project created. Then, follow these steps:
- Set your
CONVEX_DEPLOY_KEYin your local environment variables or.envfile. - Deploy the backend logic from the skill directory:
cd ~/.openclaw/skills/convex-backend
CONVEX_DEPLOY_KEY=your_key_here npx -y convex@latest deploy
- Ensure the Convex MCP server is configured in your agent's settings:
{
"mcpServers": {
"convex": {
"command": "npx",
"args": ["-y", "convex@latest", "mcp", "start"]
}
}
}
- Verify the setup using mcporter:
npx -y mcporter list convex --schema
Convex Backend Data Schema & Taxonomy
The skill organizes data into two main categories: Environment Variables and Structured Memory.
| Data Category | Tool Used | Purpose |
|---|---|---|
| Secrets | convex.envSet |
Manages API keys and tokens with agent-specific overrides. |
| Memory | memory:addMemory |
Stores facts, preferences, and decisions as structured objects. |
| Logs | memory:writeDailyLog |
Appends session summaries to a date-indexed log. |
Memory entries include metadata such as agentId, type (fact, preference, decision, note), and searchable tags to keep Openclaw Skills organized and efficient.
Convex Backend Advanced Features
- Automatic migration of local .env keys to the cloud with secure verification phases.
- Hierarchical secret resolution (Agent-specific > Shared Default > Global Fallback).
- Support for
mcporterbridging in environments without native MCP tool exposure. - Component-based architecture allowing users to add custom backend tables and functions without breaking core integration logic.
- Built-in session startup checklist for context loading and continuity.
SKILL.md
Loading
Related Openclaw Skills
GitOps Team Operations by LaunchThatBot
A role-based GitOps skill that enables AI agents to function as junior contributors or senior maintainers within GitHub repositories.
Convex Backend
Store secrets, long-term memory, and daily logs in a persistent Convex backend instead of volatile local files.
LaunchThat OpenClaw Connector
A specialized bridge for connecting OpenClaw VPS instances to LaunchThatBot with secure authentication and event reporting.
Apple Mail Management Expert
An AI-powered assistant for Apple Mail that automates inbox triage, organization, and productivity workflows to help you reach inbox zero.
launchthatbot-git-team-ops
A role-based GitOps skill that assigns junior or senior operating modes to AI agents for secure, governed GitHub repository management.
Fathom
A powerful integration for retrieving meeting intelligence, including transcripts and action items, directly from Fathom AI.
METADATA
Featured*
