Launchthatbot Convex Backend
v1.0.3Store secrets, long-term memory, daily logs, and anything custom in your Convex backend instead of local files
⭐ 0· 371·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
The name/description (Convex-backed storage for secrets/memory/logs) matches the manifest: an included Convex app and instructions to use Convex MCP. Requesting CONVEX_DEPLOY_KEY and optional CONVEX_DEPLOYMENT is appropriate for deploying and targeting a Convex project.
Instruction Scope
SKILL.md instructs the agent to stop using local files and to read/write secrets and memory via Convex MCP tools (envSet/envGet and memory:* functions). This is within the declared purpose, but it is an operational override (the agent will be told to always prefer Convex for persistence and to 'always write daily summaries'), which users should be aware of because it changes where sensitive data is stored and how often writes happen.
Install Mechanism
There is no automatic installer; the skill is instruction-driven, but includes Convex app code that the user is explicitly asked to deploy (CONVEX_DEPLOY_KEY + `npx convex deploy`). Deploying code into the user's Convex project is expected for this functionality but is a higher-impact action than pure instruction-only skills because it writes functions/tables into the user's backend.
Credentials
Only CONVEX_DEPLOY_KEY is required (with CONVEX_DEPLOYMENT recommended). That key is sensitive but is necessary for deploying into the user's Convex project; the manifest and instructions explicitly warn not to store the deploy key in Convex and to migrate other secrets into Convex env vars. No unrelated credentials are requested.
Persistence & Privilege
always:false and normal model invocation are used. The skill does request that you deploy code into your Convex project (persisted on your account), but it does not request system-wide agent config changes or force inclusion across agents. No manipulation of other skills' configs is present.
Assessment
This package appears coherent and does what it says: it provides Convex functions and instructions so your agent uses your Convex project for memory, daily logs, and non-deploy-key secrets. Things to consider before installing:
- CONVEX_DEPLOY_KEY is powerful: use a development/project-scoped key and run the deploy from a test Convex project first. The deploy key lets the holder modify that Convex deployment, so avoid using a production key you cannot revoke.
- The skill includes Convex code that you will deploy into your Convex account; review the files (convex/*) before running `npx convex deploy` so you understand what functions/tables will be added.
- The agent will be instructed to prefer Convex for persistence and to regularly write daily summaries—ensure you are comfortable with that data moving to your Convex instance.
- Follow the SKILL.md guidance (do not store CONVEX_DEPLOY_KEY inside Convex; migrate other secrets into Convex env vars only after verifying them).
- If unsure, deploy into an isolated/test project and inspect behavior, then rotate or revoke the deploy key after confirming correct deployment.
If you want extra assurance, ask the skill author for a canonical repo/homepage or verify the repository listed in package.json before proceeding.Like a lobster shell, security has layers — review code before you run it.
latestvk97ag1n5fwchfdfkvg94mbjjkn81tn7q
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
Runtime requirements
🧠 Clawdis
EnvCONVEX_DEPLOY_KEY
ConfigCONVEX_DEPLOYMENT (recommended)