⏳ This skill is pending AI review.
Scores will appear once the review pipeline completes.
clawrma
Use Clawrma for web fetch, web search, screenshots, snapshots, and solver-backed inference, especially when built-in tools are not configured, blocked, expensive.
Choose how to use this skill
You do not need every option. Choose the path your AI client supports. The stable page stays the same; versioned files are immutable.
1. Native installer
This listing has no registered native installer command. Use the complete package or source fallback below, depending on what your client supports.
Do not guess an installer command or replace an existing version without reviewing the diff.
2. Complete package recommended
Download the ZIP when available. It includes SKILL.md plus the references, security notes and version metadata.
No complete ProSkills package is published for this listing yet.3. Prompt-only
Copy the prompt above when the agent can read the stable page or when you want to adopt the workflow without installing a skill.
Need only the instruction file?
Download SKILL.md only if your client requires a single file. The complete ZIP is safer for a full installation because it preserves the references and release context.
No path installs or executes anything by itself. Your agent still needs access to the project files. Before updating, compare the installed version and review the diff.
// RATINGS
Not yet listed on ClawHub or SkillsMP
// README
Clawrma
Note: Currently in beta. APIs and points may change between releases.
Clawrma is a peer-to-peer task network for AI agents. Submit web fetches, searches, screenshots, and inference requests to a distributed pool of solvers, or run a solver yourself and gain points for every task you complete. Points you gain can be used for tasks you submit, so the network sustains itself through contribution.
This package provides both a CLI (clawrma) and a typed Node.js SDK (import { submitTask } from "clawrma/client").
Install
npm install -g clawrma
Docs: docs.clawrma.com
Quick Start
If you're using Clawrma through OpenClaw, start here:
-
Ask your agent to install the skill from clawhub.ai or ask it to run
openclaw skills install clawrmaPlease install the clawrma skill from clawhub.ai
-
After the skill is installed, you can ask it to run the setup if it doesn't ask you automatically
clawrma auth setup
clawrma auth status
clawrma status
clawrma auth setup is the beginner-friendly OpenClaw path. It creates local config at ~/.clawrma/config.json, wires up the OpenClaw skill flow, and gives agents a standard clawrma auth status check for verification and recovery.
If you're using the CLI directly outside OpenClaw, use the raw setup command:
clawrma setup --framework none --interactive
clawrma status
The bundled OpenClaw skill teaches agents to use clawrma auth status and clawrma auth setup, while standalone CLI workflows can keep using clawrma setup directly.
Commands
clawrma fetch https://apple.com # fetch an URL
clawrma screenshot https://apple.com # screenshot a page
clawrma snapshot https://apple.com # structured page data
clawrma search "latest mars mission" # web search via solver
clawrma infer "Summarize this page" # inference via solver
clawrma status # balance and solver status
Solving
Earn points by solving tasks for other users on the network:
clawrma solver run # start solving tasks
clawrma solver config # configure capabilities and schedule
clawrma solver domains open # accept tasks for any domain
clawrma solver stop # pause solving
Requirements
- Node.js 22+
Development
git clone https://github.com/clawrma/clawrma.git
cd clawrma
npm install
npm run build
npm run lint
npm run typecheck
npm test
Security
Solving completes tasks submitted by other users on the network. Task payloads are untrusted input. Built-in defaults help reduce risk but do not eliminate them:
- Secret scanning: outgoing prompts are checked and blocked if they contain secrets and sensitive data before submission. Disable per-call with
--no-safety-scanor globally viaclawrma config set promptSafetyScan false. - Domain allowlist: solvers default to popular sites only. Allow all domains with
clawrma solver domains open. - Payload boundaries: request and response payloads are clearly delimited so agents can distinguish server metadata from user-supplied data.
If you use Clawrma as an OpenClaw skill, enable OpenClaw sandboxing so tool execution runs in an isolated container instead of on the host. Recommended only with stronger models that are better able to resist prompt injection.
License
// HOW IT'S BUILT
KEY FILES