Requirements
- Target platform
- OpenClaw
- Install method
- Manual import
- Extraction
- Extract archive
- Prerequisites
- OpenClaw
- Primary doc
- SKILL.md
Tencent SkillHub Β· AI
Launchthatbot Convex Backend
Store secrets, long-term memory, daily logs, and anything custom in your Convex backend instead of local files
skill openclawclawhub Free
Store secrets, long-term memory, daily logs, and anything custom in your Convex backend instead of local files
β¬ 0 downloads β
0 stars Unverified but indexed
Install for OpenClaw
Quick setup
- Download the package from Yavira.
- Extract the archive and review SKILL.md first.
- Import or place the package into your OpenClaw setup.
Package facts
- Download mode
- Yavira redirect
- Package format
- ZIP package
- Source platform
- Tencent SkillHub
- What's included
- README.md, SKILL.md, convex/_generated/api.js, convex/_generated/server.js, convex/components/openclawBackend/README.md, convex/components/openclawBackend/convex.config.js
Validation
- Use the Yavira download entry.
- Review SKILL.md after the package is downloaded.
- Confirm the extracted package contains the expected setup assets.
Install with your agent
Agent handoff
Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.
- Download the package from Yavira.
- Extract it into a folder your agent can access.
- Paste one of the prompts below and point your agent at the extracted folder.
New install
I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Then review README.md for any prerequisites, environment setup, or post-install checks. Tell me what you changed and call out any manual steps you could not complete.
Upgrade existing
I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Then review README.md for any prerequisites, environment setup, or post-install checks. Summarize what changed and any follow-up checks I should run.
Trust & source
Release facts
- Source
- Tencent SkillHub
- Verification
- Indexed source record
- Version
- 1.0.3
Provenance
- Publisher
- launchthatbot
- Source page
- View original listing
- Canonical URL
- Open canonical page
Documentation
What is LaunchThatBot
LaunchThatBot.com is a platform for operating OpenClaw agents with a managed control plane, security defaults, and real-time visibility (including office/org chart style views) while still keeping your agents on your infrastructure. You own your agents and infrastructure, LaunchThatBot helps you manage deployments. Disconnect any time and keep your system running.
What this skill is for
@launchthatbot/convex-backend is for users who want longterm agent memory and secrets to persist in Convex (https://www.convex.dev/) instead of local files. Works for single agents or multi-agents working off one shared system. Use this skill when you want: durable memory across restarts structured daily logs safer secret handling through Convex env tools This skill can be used without any active connection to LaunchThatBot. It uses the stock Convex MCP server with your own Convex credentials and writes memory/logs and env-managed secrets into your Convex instance.
Manual setup required
This skill requires manual Convex setup by the user: Create a Convex account and project. From the Convex dashboard, copy the Development deploy key for that project (for now). Give that key to the bot when asked, or set it manually as local CONVEX_DEPLOY_KEY in .env / runtime env vars. Do not assume this key exists automatically. If it is missing, stop and ask the user to complete the manual setup first. You do not need to keep npx convex dev running for this skill flow. However, Convex MCP and CLI calls still need deployment context: CONVEX_DEPLOY_KEY (required for auth) CONVEX_DEPLOYMENT (or equivalent project/deployment selection via env/config) If those values are not available, stop and ask the user to set them first (for example in .env / .env.local or via MCP startup config).
Convex Backend Integration
You are connected to a Convex backend via the Convex MCP server. This gives you persistent, cloud-based storage that survives container restarts and can be accessed from the LaunchThatBot dashboard.
MCP Requirement (Stock Convex MCP)
This skill uses the stock Convex MCP server (convex@latest mcp start), not a custom LaunchThatBot MCP tool for data storage. Recommended setup: { "mcpServers": { "convex": { "command": "npx", "args": ["-y", "convex@latest", "mcp", "start"] } } } Reference: https://docs.convex.dev/ai/convex-mcp-server
mcporter Compatibility Preflight
If your runtime does not expose MCP tools natively (common in OpenClaw/Pi flows), use mcporter as the bridge. Check prerequisites before running this skill workflow: mcporter --version || npx -y mcporter --version mcporter list || npx -y mcporter list mcporter list convex --schema || npx -y mcporter list convex --schema If mcporter is not runnable, tell the user they need npx mcporter (or a local install). If convex MCP is missing, ask the user to configure stock Convex MCP, then re-run the checks. When bridging via mcporter, call Convex MCP tools through CLI commands (for example npx -y mcporter call convex.run ...) instead of assuming native MCP tool invocation in the runtime. Convex MCP tool mapping for this skill: Use convex.run for deployed Convex functions such as memory:addMemory, memory:searchMemory, memory:writeDailyLog, memory:getDailyLog, and memory:listDailyLogs. Use convex.envSet, convex.envGet, convex.envList, and convex.envRemove for environment variables. Do not call memory:* as MCP tools directly. They are function names passed to convex.run.
CRITICAL RULES
CONVEX_DEPLOY_KEY is the only required local secret in .env. CONVEX_DEPLOYMENT may also be set locally for explicit deployment targeting (and can be derived from the deploy key prefix). NEVER store CONVEX_DEPLOY_KEY in Convex (envSet). All other secrets (API keys, tokens, passwords) must use Convex environment variables (envSet/envGet/envList/envRemove) and not local .env or auth.json. NEVER store sensitive credentials in MEMORY.md or daily log files. Always write daily summaries to Convex at the end of each session or when asked to save progress. Always check Convex for existing secrets before asking the user to provide them again.
Required local credentials
Before running this skill, the user must provide: CONVEX_DEPLOY_KEY (required local secret) CONVEX_DEPLOYMENT (recommended for deterministic targeting; may be parsed from deploy key prefix) If deployment auth/targeting is unclear, stop and request user confirmation before writes.
Bootstrap Rule (Required)
Before Convex MCP operations can work, CONVEX_DEPLOY_KEY must already exist in local runtime env (.env / deployment env vars). CONVEX_DEPLOYMENT should also be set when available for deterministic targeting. If CONVEX_DEPLOY_KEY is missing: stop secret/memory operations tell the user exactly what is missing ask them to set CONVEX_DEPLOY_KEY locally (and set CONVEX_DEPLOYMENT by parsing the deploy key prefix when possible), then restart/reload MCP re-run preflight: mcporter list convex --schema || npx -y mcporter list convex --schema Do not attempt to save CONVEX_DEPLOY_KEY to Convex.
Required Convex Deploy Step
This skill package includes a Convex app under the convex/ folder and convex.json at package root. The required backend logic is mounted as a local component: component name: convex_openclaw_backend_component app mount alias: openclawBackend After installing or updating this skill, run deploy from the skill root: cd /home/node/.openclaw/skills/convex-backend CONVEX_DEPLOY_KEY=... npx -y convex@latest deploy Why: npx convex deploy must run in the directory containing convex.json. Without deploy, the custom memory/daily-log functions are not available, and memory:* calls will fail.
Rule for Adding Custom Logic
If the user asks for new backend functionality (for example custom task tables, new mutations, or queries): Keep required integration logic inside component files (convex/components/openclawBackend/*). Put user-specific custom tables/functions in the root app (convex/*) so they stay separate from required integration tables. Re-run deploy from skill root each time changes are made. While autonomously adding custom logic, never modify or delete the openclawBackend component. Treat openclawBackend as core integration infrastructure, not a workspace for user-specific features. If you change Convex code but do not deploy, newly added functions/tables will not exist in the user's Convex deployment.
Storing Secrets (API Keys, Passwords, Tokens)
Use stock Convex MCP environment variable tools for secrets: envSet envGet envList envRemove Do not use custom secrets:* functions for credentials in this skill.
Secret naming strategy (shared + per-agent fallback)
For a logical secret key like OPENAI_API_KEY, resolve in this order: AGENT_<agentId>_OPENAI_API_KEY (agent-specific override) AGENT_DEFAULT_OPENAI_API_KEY (shared default for all agents) OPENAI_API_KEY (legacy global fallback, optional) Examples: Agent override: AGENT_agent2_OPENAI_API_KEY Shared default: AGENT_DEFAULT_OPENAI_API_KEY
Write / Read / Remove examples
Set shared default: Tool: envSet Arguments: { "name": "AGENT_DEFAULT_OPENAI_API_KEY", "value": "sk-..." } Set agent-specific override: Tool: envSet Arguments: { "name": "AGENT_<agentId>_OPENAI_API_KEY", "value": "sk-..." } Read by fallback chain: envGet("AGENT_<agentId>_OPENAI_API_KEY") if missing, envGet("AGENT_DEFAULT_OPENAI_API_KEY") if missing, optionally envGet("OPENAI_API_KEY") Remove an agent override: Tool: envRemove Arguments: { "name": "AGENT_<agentId>_OPENAI_API_KEY" }
First-Run Migration for Existing .env Keys
If this skill is installed on an existing agent that already has many keys in local .env, run this migration prompt after Convex MCP preflight succeeds: Ask the user: "Convex backend is configured. Do you want me to migrate all local .env secrets into Convex and remove them from local .env? Recommended: Yes. Local .env will keep CONVEX_DEPLOY_KEY (required) and optionally CONVEX_DEPLOYMENT for explicit targeting." If user confirms, run migration in two phases:
Phase A: Copy + verify (non-destructive)
Read local .env and collect secret key/value pairs. Exclude CONVEX_DEPLOY_KEY and CONVEX_DEPLOYMENT. For each remaining key, copy to Convex env using naming convention above: preferred: AGENT_DEFAULT_<KEY> optional per-agent override: AGENT_<agentId>_<KEY> Verify migration with envList and targeted envGet. Report copied-key count and ask for explicit cleanup approval.
Phase B: Optional cleanup (destructive)
Only continue if the user explicitly confirms cleanup with a clear yes response (for example: YES_REMOVE_LOCAL_ENV). Remove only keys that were copied and verified in Phase A. Keep CONVEX_DEPLOY_KEY in local .env, and optionally keep CONVEX_DEPLOYMENT for explicit deployment targeting. Confirm cleanup completion with removed-key count. Safety notes: Create a local backup of .env before mutation. Do not print secret values in chat/log output. If any key fails migration, do not delete that key from local .env until retry succeeds.
Storing Long-Term Memory
When you learn something important about the user, their preferences, or make a significant decision: Tool: convex.run (via mcporter) Function: memory:addMemory Arguments: { "agentId": "<your-agent-id>", "type": "fact", "content": "User prefers TypeScript over JavaScript for all new projects", "tags": ["preferences", "coding"] } npx -y mcporter call convex.run --args '{ "functionName": "memory:addMemory", "args": { "agentId": "<your-agent-id>", "type": "fact", "content": "User prefers TypeScript over JavaScript for all new projects", "tags": ["preferences", "coding"] } }' Memory types: fact β Something true about the user or their setup preference β User likes/dislikes decision β A choice that was made and should be remembered note β General observations or context To recall memories: Tool: convex.run (via mcporter) Function: memory:searchMemory Arguments: { "agentId": "<your-agent-id>", "type": "preference", "limit": 20 }
Daily Log Entries
At the end of each work session, write a summary of what was accomplished: Tool: convex.run (via mcporter) Function: memory:writeDailyLog Arguments: { "agentId": "<your-agent-id>", "date": "2026-02-17", "content": "## Summary\n- Set up email integration with Resend\n- Configured GitHub SSH keys\n- Started work on Twitter bot automation\n\n## Blockers\n- Need Twitter API key from user" } Daily logs are append-only β calling writeDailyLog for the same date appends to the existing entry. To review past logs: Tool: convex.run (via mcporter) Function: memory:listDailyLogs Arguments: { "agentId": "<your-agent-id>", "limit": 7 }
Session Startup Checklist
At the beginning of each session: Check for configured env secrets: convex.envList (and convex.envGet for required keys) Load recent memories: convex.run with function memory:searchMemory and limit 20 Load today's log: convex.run with function memory:getDailyLog and today's date Load yesterday's log for continuity context This ensures you have full context from previous sessions.
Your Agent ID
Your agent ID is provided in your agent configuration. Use it consistently in all Convex calls. If you're unsure of your agent ID, check your agent YAML config file.
Category context
Agent frameworks, memory systems, reasoning layers, and model-native orchestration.
Source: Tencent SkillHub
Largest current source with strong distribution and engagement signals.
Package contents
Included in package
3 Docs3 Scripts
- SKILL.md
- convex/components/openclawBackend/README.md
- README.md
- convex/_generated/api.js
- convex/_generated/server.js
- convex/components/openclawBackend/convex.config.js