CashClaw skill
Thirteen skills turn an OpenClaw agent into a worker for paid jobs, while Guard caps its spending and stops loops
- 306 ★ on the repository
- Last commit
- MIT
Broad access
Why this level
- The skills invoice through Stripe, take payment and withdraw funds to the worker's wallet
- Auto-accept takes jobs without human confirmation, and the skills send emails and messages to clients and work with WhatsApp
- The data scraping skill crawls sites and compiles contact lists, and Stripe and HYRVE keys sit in the configuration
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/cashclawDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewBroad accessThe tool can change data, run commands or handle secrets.
- The skills invoice through Stripe, take payment and withdraw funds to the worker's wallet
- Auto-accept takes jobs without human confirmation, and the skills send emails and messages to clients and work with WhatsApp
- The data scraping skill crawls sites and compiles contact lists, and Stripe and HYRVE keys sit in the configuration
This is third-party code. Review the repository files before installing.
What it does
CashClaw turns an OpenClaw agent into a worker for paid digital jobs. Its 13 skills cover typical services, from SEO audits and copywriting to cold email and WhatsApp work. The core skill, cashclaw-core, runs the job lifecycle, client correspondence and revenue tracking, and cashclaw-invoicer issues invoices and payment links through Stripe. The missions folder holds 17 job templates with a scope of work. `npx cashclaw init` creates a ~/.cashclaw/ workspace, sets up the job pipeline, installs the skills into the agent and registers it on the HYRVE AI marketplace. With the cashclaw hyrve commands the agent finds jobs, accepts them, delivers work and checks its profile. An auto-accept mode takes proposals under a set amount with no human in the loop. A separate Guard skill protects against runaway spending. A YAML policy sets a cost ceiling per call and per day, a limit on repeats of the same request and a denylist of tools such as shell and eval. On a breach Guard throws BudgetExceeded or RecursionKilled.
Who it is for. For people trying OpenClaw agents as freelancers, and developers who need spend and repeat limits for autonomous agents.
Good fit when
- You want to try the idea of an agent that takes SEO audit, copywriting or lead list jobs and invoices through Stripe
- An autonomous agent needs a hard spend ceiling on model calls so a misconfiguration cannot produce a bill of thousands of dollars
- You need to stop an agent stuck on the same request and get an alert in Telegram or Slack
- You want to forbid dangerous tools and cap message rates with a YAML policy
Not a fit when
- You do not use OpenClaw: the skills follow its format and install with `cashclaw init`
- The agent must not touch payments and client correspondence without human review
- You want a library for cost reporting without hard limits: Guard works as a ceiling that interrupts the call
- Outreach or data collection has no legal basis or recipient consent: the skills only carry out jobs and remove no responsibility
Example request
Run a standard-tier SEO audit of the client's site, write the report and prepare a Stripe invoice for my approvalLimitations
Node.js is required. Accepting payments needs a Stripe account, and the marketplace needs a HYRVE AI key. The marketplace and its commission belong to a separate service, and the worker receives 85 percent of the payment. Payouts go through Stripe or USDT.
The Guard policy sits in guard-policy.yaml inside the workspace, and in code model calls are wrapped through the SDK, for example guard.llm. Payments over the Machine Payments Protocol in USDC arrived in version 1.7.0, so test them with small amounts. The results and quotes in the repository come from the author and have no independent check.
How to disable. Delete the `cashclaw-*` folders from the OpenClaw skills directory and run `npm uninstall -g cashclaw`. The ~/.cashclaw/ workspace can be deleted by hand, and Stripe and HYRVE keys should be revoked with their owners.
FAQ
How do I install CashClaw for an OpenClaw agent?
Run `npx cashclaw init`, or install the package globally with `npm install -g cashclaw` and run `cashclaw init`. The wizard walks you through the steps, and Stripe and the marketplace can be skipped.
What does CashClaw Guard do?
It stops calls that fall outside the policy. Cost ceilings per call and per day, a token limit, a repeat limit for one request and a tool denylist are set in YAML. The commands `cashclaw guard init`, `status`, `test`, `kill`, `logs` and `reload` manage the policy, and alerts go to Telegram, Slack or Discord.
How do I turn off job auto-accept?
With `cashclaw hyrve auto-accept off`. It is turned on with `cashclaw hyrve auto-accept on` and an amount cap such as `--max 500`, and then accepts proposals under the cap without human confirmation.
Which services do the missions templates cover?
17 templates: blog posts of 500 and 1,500 words, competitor analysis, data scraping, cold email, landing pages, a 50-lead list, a reputation audit, an SEO audit, a week of social posts and WhatsApp setup. Many come in a basic and an extended tier.
Do I need Stripe to use it?
No, Stripe is optional and can be connected at any time. Without it, invoices and payment links in the cashclaw-invoicer skill do not work.
Discussion
No questions or reviews yet.
Related
SkillsA self-hosted control plane for running several agents: task dispatch, run review, spend tracking, connected via MCP, CLI or REST
Yandex Smart Home for OpenClaw
Yandex Smart Home CLI for OpenClaw
Python CLI and OpenClaw skill to switch Yandex Smart Home lights and sockets and set brightness, color temperature and color
Orchestrator for Claude Code, Codex and Gemini CLI: YAML workflows that loop until your own quality gates pass
Bitrix24 as mission control for AI dev agents: a kanban task conveyor, shared memory and one-button phone acceptance
Skills for coding agents that build and verify workspaces on the Lemma platform, with their agents and apps
1688 Shopkeeper
1688-shopkeeper
The official 1688 skill for OpenClaw: supplier search, product details, listing to Douyin, Pinduoduo, Xiaohongshu and Taobao, trends and a daily report
Comparison with similar tools
Entries that share the main search query, and how they differ.
| Entry | Official | Repository stars | Main difference |
|---|---|---|---|
| CashClaw (this entry) | no | 306 | |
| Cold Outbound Skills | no | 792 | 31 skills on cold email only, with no invoicing, job marketplace or spend limits |
| SEOmator SEO Audit | no | 654 | a single skill for a full technical SEO audit of a site |
README in short
The README frames the set as a layer of an agent economy where agents earn and spend while Guard watches the costs. A quickstart, a diagram linking OpenClaw with Stripe and HYRVE AI, and a sample Guard policy follow.
README badge
Original SKILL.md
---
name: cashclaw-core
description: The business brain of CashClaw. Orchestrates mission lifecycle, client communication, revenue tracking, and delegates work to specialized skills.
metadata:
{
"openclaw":
{
"emoji": "\U0001F99E",
"requires": { "bins": ["node", "cashclaw"] },
"install":
[
{
"id": "npm",
"kind": "node",
"package": "cashclaw",
"bins": ["cashclaw"],
"label": "Install CashClaw via npm"
}
]
}
}
---
# CashClaw Core - Business Orchestration Engine
You are the CashClaw business brain. Your sole purpose is to turn AI capabilities into
revenue by managing the full lifecycle of paid missions. Every interaction you have should
move a mission forward or generate a new one.
## Mission Lifecycle
Every paid engagement follows this exact 8-stage pipeline. Never skip a stage.
### Stage 1: INTAKE
When a new client request arrives:
1. Parse the client message for: service type, scope, deadline, budget hints.
2. Create a mission file at `~/.cashclaw/missions/MISSION-{YYYYMMDD}-{SEQ}.md`.
3. Log the intake in `~/.cashclaw/ledger.jsonl` with status `intake`.
4. Identify which CashClaw skill(s) are needed.
5. Ask clarifying questions if scope is ambiguous. Never assume; always confirm.
Mission file template: