A protocol for ClawWork, an economic-survival simulation where an AI agent earns money by completing tasks and spends money on each API call.

Use it to follow the simulation’s daily cycle, check financial status, learn, submit work, and create or read supported files.

It makes the agent manage its balance by choosing between paid work and unpaid learning while meeting a quality threshold.

clawwork — Skill for Claude Code and Codex by HKUDS

About the project

ClawWork is a system for evaluating AI assistants as coworkers by having them complete professional tasks, earn income, cover their own token costs, and maintain economic solvency. It is intended for testing how AI agents perform on real-world work across multiple professions.

HKUDS/ClawWork · 8,566 stars · on GitHub

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent

npx skills add HKUDS/ClawWork --skill skill

Clone the repo

git clone --depth 1 https://github.com/HKUDS/ClawWork

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/hkuds/clawwork/skill/github.svg)](https://agentmods.dev/skills/hkuds/clawwork/skill)

Your own site

<a href="https://agentmods.dev/skills/hkuds/clawwork/skill"><img src="https://agentmods.dev/badge/skills/hkuds/clawwork/skill/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/hkuds/clawwork/skill"><img src="https://agentmods.dev/badge/skills/hkuds/clawwork/skill.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 16 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 585 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits

NVIDIA SkillSpector pass 7 Sept 2026

How audits are shown

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00016	$0.00585
Opus 5.5	$0.00006	$0.00234
Sonnet 5.5	$0.00003	$0.00117
Haiku 4.5	$0.00002	$0.00059

Measured 2mo ago against content hash 506b64fb43cf, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

clawwork scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2mo ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

clawmode_integration/skill/SKILL.md · 56 lines

How it starts

The opening of the file, as written. The whole thing — 56 lines — stays where its author put it; the contents beside it link to each section on GitHub.

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

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

Available Economic Tools

Tool	Purpose
`decide_activity`	Choose "work" or "learn" for today (required first step)
`submit_work`	Submit text and/or file artifacts for evaluation and payment
`learn`	Save knowledge to persistent memory (min 200 chars)
`get_status`	Check balance, net worth, and survival status
`create_artifact`	Create a work artifact file (txt, md, csv, json, xlsx, docx, pdf)
`read_artifact`	Read a file and return its content (pdf, docx, xlsx, pptx, png, jpg, txt)

Daily Workflow (Benchmark Mode)

Analyse your economic status (already shown in the prompt — do NOT call get_status redundantly).
Decide: call decide_activity(activity="work", reasoning="...").
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="...").
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.

Read the full file on GitHub · 56 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

2mo ago First seen · 56 lines · 16 tokens per session scan A 506b64fb43cf

Subscribe to this mod's changes

clawwork is a skill published in the GitHub repository HKUDS/ClawWork (8,566 stars, last pushed 8mo ago), licensed MIT. It adds 16 tokens to every session and 585 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

clawwork next to 6 skills from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
clawworkHKUDS/ClawWork	Skill	8.6k	8mo ago	MIT	A	16	Claude Code, Codex
imagegenopenai/codex	Skill	133k	yesterday	Apache-2.0	A	113	Codex
playwright-climicrosoft/playwright	Skill	99k	yesterday	Apache-2.0	A	19	Claude Code
local-ai-agentsmicrosoft/ai-agents-for-beginners	Skill	79k	4d ago	MIT	A	200	Claude Code, Codex
systematic-debuggingobra/superpowers	Skill	313k	3d ago	MIT	A	21	Claude Code
next-cache-components-adoptionvercel/next.js	Skill	143k	yesterday	MIT	A	95	Claude Code
next-partial-prefetching-adoptionvercel/next.js	Skill	143k	yesterday	MIT	A	103	Claude Code