U-Claw is a portable AI workspace that places an OpenClaw assistant, its configuration, memory, sessions, and tools on a USB drive. Users set it up on supported computers and carry the workspace between them, configuring a model with their own API key. The catalogue contains skills and instructions related to using or preparing this portable setup.
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.
npx skills add dongsheng123132/u-claw --skill deepseek-helpergit clone --depth 1 https://github.com/dongsheng123132/u-clawWrote 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.
[](https://agentmods.dev/skills/dongsheng123132/u-claw/deepseek-helper)<a href="https://agentmods.dev/skills/dongsheng123132/u-claw/deepseek-helper"><img src="https://agentmods.dev/badge/skills/dongsheng123132/u-claw/deepseek-helper/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.
<a href="https://agentmods.dev/skills/dongsheng123132/u-claw/deepseek-helper"><img src="https://agentmods.dev/badge/skills/dongsheng123132/u-claw/deepseek-helper.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00025 | $0.00701 |
| Opus 5 | $0.00013 | $0.00351 |
| Sonnet 5 | $0.00005 | $0.00140 |
| Haiku 4.5 | $0.00003 | $0.00070 |
Grade A, and why
deepseek-helper scanned grade A with 1 finding 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 11d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- **API 调用示例**: 提供 Python、curl、Node.js 等语言的调用代码 What it actually says
DeepSeek API 助手
帮助用户高效使用 DeepSeek 系列模型，包括模型选择建议、API 调用示例、编程辅助和定价计算。
功能概述
- 模型选择指南: deepseek-v4-flash vs deepseek-v4-pro 的使用场景对比
- API 调用示例: 提供 Python、curl、Node.js 等语言的调用代码
- 定价计算: 根据 token 用量估算费用（输入 ¥1/百万tokens，输出 ¥2/百万tokens）
- Prompt 优化: 针对 DeepSeek 模型特性优化提示词
- 错误排查: 常见 API 错误码解读和解决方案
模型对比
| 模型 | 适用场景 | 上下文长度 |
|---|---|---|
| deepseek-v4-flash | 日常对话、文案写作、知识问答，快且便宜 | 128K |
| deepseek-v4-pro | 复杂推理、数学、代码生成与审查 | 128K |
⚠️ 旧名
deepseek-chat/deepseek-coder/deepseek-reasoner已下线， api.deepseek.com 现在只接受deepseek-v4-flash和deepseek-v4-pro， 用旧名调用会直接返回 HTTP 400。
使用示例
基础 API 调用
帮我写一个调用 DeepSeek API 的 Python 脚本，实现多轮对话
模型选择
我要做一个客服机器人，应该用 deepseek-v4-flash 还是 deepseek-v4-pro？
费用估算
我每天大约处理 1000 条用户消息，平均每条 500 token，帮我算下月费用
错误排查
调用 DeepSeek API 返回 429 错误，怎么处理？
适用场景
- 首次接入 DeepSeek API 的开发者
- 需要在 deepseek-v4-flash 和 deepseek-v4-pro 之间做选择
- 想要优化 API 调用成本
- 排查 API 调用问题
- 需要快速生成 API 调用代码片段
不适用场景
- 需要调用 OpenAI / Claude 等其他模型的 API（请使用对应工具）
- 需要实时获取 DeepSeek 最新公告（本技能基于已知信息）
- 需要部署私有化 DeepSeek 模型（涉及运维，超出范围）
关键提示
- DeepSeek API 兼容 OpenAI 格式，base_url 改为
https://api.deepseek.com - 国内访问无需翻墙，延迟低
- 支持 function calling 和 JSON mode
- 建议使用 stream 模式提升用户体验
- API Key 申请地址: https://platform.deepseek.com
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.
- 11d ago First seen · 73 lines · 25 tokens per session scan A d75759a05d90
deepseek-helper is a skill published in the GitHub repository dongsheng123132/u-claw (1,747 stars, last pushed 3d ago), licensed MIT. It adds 25 tokens to every session and 701 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
proxy-local-ai-subscriptions
A guide for exposing your local Codex, ChatGPT Codex, or Claude Code subscription through a protected local OpenAI-compatible endpoint, then connecting it to NextClaw as a custom provider.
opik-optimizer
Optimize LLM prompts, tools, and agents in Opik using standardized optimizer workflows (prompt optimization, tool optimization, and parameter tuning), dataset/metric wiring, and result interpretation.
llm-engineering-expert
Build reliable applications on large language models: prompt design, structured output, evaluation, guardrails, and cost and latency control. Use when the user mentions LLMs, prompts, prompt engineering, few-shot examples, structured or JSON output, function calling, hallucination, model evaluation, token costs…
openai-docs
Use when the user asks how to build with OpenAI products or APIs and needs up-to-date official documentation with citations, help choosing the latest model for a use case, or model upgrade and prompt-upgrade guidance; prioritize OpenAI docs MCP tools, use bundled references only as helper context, and restrict any…
building-agent-systems
AI agent and LLM system engineering reference covering single-agent dev (ReAct, tool calling, plan-execute), multi-agent coordination (swarm, role decomposition, file locking), LLM security (prompt injection, jailbreak defense, output filtering), RAG architecture (chunking, hybrid retrieval, rerank), and prompt…
ml
Machine learning and LLM engineering judgment, distilled from a stronger model - invoke when DECIDING whether/how to use ML or an LLM for a task (prompt vs RAG vs fine-tune vs classical); working with training/eval data or labels; building or reviewing evals for models and LLM features; designing RAG, structured…