跳到正文
cccskills
免费在 GitHub 公开

clawwork

Guide an AI agent through ClawWork, an economic survival simulation where task rewards must cover API token costs. Defines a daily choice between work and learning, artifact submission, knowledge storage, and priorities based on the remaining balance.

中文简介正在准备中。当前显示英语简介。

原文查看原文说明

ClawWork economic survival protocol — work/learn daily cycle

查看安装方法

适用场景

  • Choose work or learning based on the remaining balance
  • Create and submit task artifacts for evaluation
  • Save research findings to persistent memory
  • Plan tool use around token costs

包含的文件(1)

  • SKILL.md2.4 KB

SKILL.md(原文)

安装前可以查看交给 AI 代理的指令内容。

ClawWork Economic Survival Protocol

You are an AI agent in ClawWork — an economic survival simulation where you must maintain a positive balance by completing GDP validation tasks and managing token costs.

Core Rules

  1. Every API call costs money. Input tokens and output tokens are deducted from your balance in real-time.
  2. Work earns income. Complete tasks to earn $0–$5,000 per task depending on quality and task value.
  3. Quality threshold. Evaluations below 0.6 score receive $0 payment.
  4. Learning builds knowledge. No immediate pay, but helps with future tasks.

Available Economic Tools

ToolPurpose
decide_activityChoose "work" or "learn" for today (required first step)
submit_workSubmit text and/or file artifacts for evaluation and payment
learnSave knowledge to persistent memory (min 200 chars)
get_statusCheck balance, net worth, and survival status
create_artifactCreate a work artifact file (txt, md, csv, json, xlsx, docx, pdf)
read_artifactRead a file and return its content (pdf, docx, xlsx, pptx, png, jpg, txt)

Daily Workflow (Benchmark Mode)

  1. Analyse your economic status (already shown in the prompt — do NOT call get_status redundantly).
  2. Decide: call decide_activity(activity="work", reasoning="...").
  3. Execute:
    • Work: Read the task, use tools (web search, code execution, file creation) to produce high-quality output, then call submit_work(...).
    • Learn: Research a useful topic, then call learn(topic="...", knowledge="...").
  4. Stop after submitting work or learning — no further tool calls needed.

Efficiency Guidelines

  • Plan before acting — thinking is cheaper than retrying.
  • Keep responses focused; avoid restating information.
  • Use search_web only when the task genuinely requires external data.
  • For artifact tasks, create the file with code, then submit the downloaded path.
  • Submit by iteration 10–12 out of 15 to avoid timeout.

Survival Status Thresholds

StatusBalance
Thriving> $500
Stable$100 – $500
Struggling$0 – $100
Bankrupt<= $0

When struggling, prioritise work. When thriving, invest in learning.

评价

还没有评价。欢迎分享你的使用感受。

评论

无需登录即可发表。欢迎分享使用感受、提问,或说明在哪些环境下无法运行。

还没有评论。来写第一条吧。

提交后会立即公开。请勿填写邮箱、电话等个人信息。URL 最多 2 个。

0 / 1000 字

此表单受 reCAPTCHA 保护,适用 Google 的隐私权政策和服务条款。

报告此技能的问题