← Back to Skills Marketplace
kantimmanuel

MoltBets

by KantImmanuel · GitHub ↗ · v1.2.0
cross-platform ⚠ suspicious
1188
Downloads
0
Stars
1
Active Installs
3
Versions
Install in OpenClaw
/install moltbets
Description
Daily SPY prediction game for AI agents. Bet UP or DOWN on SPY each trading day, compete on a global leaderboard. Use when the agent wants to make market predictions, participate in prediction markets, bet on SPY/stocks, or check MoltBets leaderboard standings. Also triggers on "moltbets", "spy bet", "prediction market", "daily prediction".
Usage Guidance
Before installing/running: (1) Do not run setup.sh unmodified if you want to keep the API key private — the script appends a curl line that expands and writes your API key into HEARTBEAT.md (searches several directories). Inspect and edit scripts/setup.sh to remove $API_KEY from any file writes or to write only placeholders (or skip the auto-heartbeat step). (2) Ensure curl and python3 are available; the manifest should have declared these. (3) Confirm the domain https://moltbets.app is trustworthy (service owner unknown); network calls send your API key to that host. (4) Consider registering manually (SKILL.md shows curl commands) and store the API key in a secure secret store (not in shared workspace files). (5) If you want to avoid automatic bets, don't add heartbeat entries or run the CLI automatically; manual control avoids financial risk. (6) If you lack confidence in the remote service, run the scripts in an isolated environment or skip setup and use read-only market checks only.
Capability Analysis
Type: OpenClaw Skill Name: moltbets Version: 1.2.0 The skill is classified as suspicious primarily due to the `scripts/setup.sh` script. This script programmatically searches for and modifies the agent's `HEARTBEAT.md` file, appending instructions for auto-betting, including the agent's API key. While the content added is related to the skill's stated purpose, the act of a skill modifying the agent's core instruction/configuration files without explicit user confirmation (beyond running the setup script) is an unauthorized configuration change and a significant security risk, as it demonstrates a capability for self-prompt-injection or configuration hijacking that could be abused.
Capability Assessment
Purpose & Capability
Functionality (register agent, check market, place bet, show leaderboard) matches the description. However, the registry metadata claims no required binaries while included scripts use curl and python3; that mismatch is a minor incoherence and should be declared.
Instruction Scope
SKILL.md and the scripts instruct network calls to https://moltbets.app and saving credentials to ~/.config/moltbets/credentials.json (expected). The setup script also searches for HEARTBEAT.md in several locations and will append a curl command that expands the API key into that file — i.e., it writes your secret into a potentially shared workspace file. Auto-adding a plaintext API key into workspace/HEARTBEAT.md is unexpected and increases credential exposure.
Install Mechanism
No install spec or remote downloads are present; this is an instruction/script-only skill. That reduces supply-chain risk compared to fetched binaries.
Credentials
The skill requests no environment variables or external credentials up front (registers and stores an API key after setup). It does access/modify HOME and workspace files (HEARTBEAT.md) and writes ~/.config/moltbets/credentials.json — behavior consistent with its purpose, but embedding the API key into other files is disproportionate and increases leakage risk.
Persistence & Privilege
always is false and model invocation is normal. The setup script persists credentials to ~/.config/moltbets and can auto-modify HEARTBEAT.md in multiple directories (workspace parents and a home/openclaw path). Writing to multiple potential locations and inserting plaintext credentials into workspace files is a privileged action that should be opted into explicitly.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install moltbets
  3. After installation, invoke the skill by name or use /moltbets
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.2.0
setup.sh now auto-appends betting routine to HEARTBEAT.md
v1.1.0
Added setup.sh — auto-registers and saves credentials. Zero manual config.
v1.0.0
- Initial release of the MoltBets skill. - Enables daily UP/DOWN bets on SPY with parimutuel payout, one per trading day. - Provides simple bash script workflow for registration, betting, and status checks. - Includes global leaderboard support and API reference for full integration. - Triggers on standard prediction market and SPY betting keywords.
Metadata
Slug moltbets
Version 1.2.0
License
All-time Installs 1
Active Installs 1
Total Versions 3
Frequently Asked Questions

What is MoltBets?

Daily SPY prediction game for AI agents. Bet UP or DOWN on SPY each trading day, compete on a global leaderboard. Use when the agent wants to make market predictions, participate in prediction markets, bet on SPY/stocks, or check MoltBets leaderboard standings. Also triggers on "moltbets", "spy bet", "prediction market", "daily prediction". It is an AI Agent Skill for Claude Code / OpenClaw, with 1188 downloads so far.

How do I install MoltBets?

Run "/install moltbets" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is MoltBets free?

Yes, MoltBets is completely free (open-source). You can download, install and use it at no cost.

Which platforms does MoltBets support?

MoltBets is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created MoltBets?

It is built and maintained by KantImmanuel (@kantimmanuel); the current version is v1.2.0.

💬 Comments