Search results for "VA"

Seko, SenseTime Seko, AI short-form drama generation, AI video creation platform, integrated creation agent, one-person production crew, AI motion comic production, Seko Infinite Canvas, Seko 2.0, SekoIDX, SekoTalk, AI-driven film & TV industrialization, AI short-form drama tools, AI animation production, SenseTime

2026-06-28

FlagEval: The internationally authoritative large model evaluation system and Libra open platform

223

FlagEval (Libra) is a large-scale model evaluation system and open platform initiated by Beijing Zhiyuan Artificial Intelligence Research Institute, aiming to establish scientific, fair, and open evaluation benchmarks and methods. The platform has innovatively constructed a three-dimensional

AGI Eval: A large model evaluation community and authoritative third-party evaluation platform

182

AGI Eval is a large model evaluation community jointly created by top universities and institutions such as Shanghai Jiao Tong University, Tongji University, East China Normal University, and DataWhale, with the mission of "assisting evaluation and making AI a better partner for humanity". The

OpenCompass: An Open Source Large Model Comprehensive Evaluation System and Sinan Open Platform

177

OpenCompass is an open-source large model evaluation system launched by Shanghai Artificial Intelligence Laboratory, providing one-stop evaluation services for large language models, multimodal models, and scientific intelligence models. The platform supports one click distributed evaluation of

AIPRM: The preferred prompt word management tool and community driven prompt word library for

171

AIPRM is a browser extension and prompt management tool for mainstream AI platforms such as ChatGPT, Claude, Gemini, Midjourney, etc. It provides over 5400 community approved ready to use prompt words. It supports guided creation of prompt words, multivariable settings, customization of tone and

CMMLU: Authoritative Chinese Language Model Knowledge Understanding Ability Evaluation Benchmark

167

CMMLU (Chinese Massive Multitask Language Understanding) is a large-scale language understanding benchmark designed specifically for the Chinese language context, covering 67 subject topics from beginner to advanced professional levels, including natural sciences, social sciences, engineering

Tip Engineering Guide: A Systematic Learning Roadmap from Zero Foundation to AI Dialogue Master

151

The Tip Engineering Guide is the world's most popular open-source tip engineering learning project initiated by the renowned AI research institution DAIR.AI. It has accumulated over 30000 stars on GitHub and covers the entire knowledge system from basic principles to advanced technologies. The

PromptPilot: A professional prompt optimization and automation tuning platform for AI instruction

145

PromptPilot is a systematic prompt word engineering platform launched by Volcano Engine, aimed at addressing the pain point of "how to efficiently ask AI questions". It provides full lifecycle management covering Prompt generation, debugging, evaluation, and iteration, supports one click rewriting

MMLU: The International Authoritative Benchmark for Multi Task Language Understanding Ability of

141

MMLU (Massive Multitask Language Understanding) is a large-scale multi task language understanding evaluation dataset jointly released by the University of California, Berkeley and other institutions. It covers 57 disciplinary fields, including humanities, social sciences, natural sciences,

HELM: Stanford University led comprehensive evaluation framework for large language models and high

131

HELM (Holistic Evaluation of Language Models) is a comprehensive language model evaluation framework initiated by the Stanford University Center for Fundamental Model Research (CRFM), aimed at systematically evaluating large language models through multidimensional, standardized, and reproducible

Ginlix – an "AI construction team" tailored for Wall Street, enabling investment analysis to be continuously iterated like writing code.

128

Ginlix is an AI-powered company specializing in financial investment research. Its flagship product, LangAlpha, is an open-source investment research framework. Employing a Programmatic Tool Call (PTC) architecture, it allows AI agents to process financial data by writing and executing code, with a persistent workspace that enables research to accumulate like a codebase. It supports built-in models and can be deployed privately. It is suitable for individuals and institutions looking to leverage AI for deep, verifiable investment decisions.

Magic Design: Canva's own AI design tool, can it really let you "generate images just by speaking a sentence"?

124

Magic Design is an AI-powered design tool from Canva that automatically generates designs—such as posters and presentations—based on text input or uploaded images. A free version is available, while the Pro version unlocks additional features.

2026-06-29

A Quick Take: Just how much hassle can that AI meeting note-taker hidden inside Baidu Netdisk really save you?

120

"Jiandan Tingji" is an AI-powered speech-to-text tool launched by Baidu Netdisk. Integrated with the ERNIE Bot (Wenxin Yiyan) large language model, it supports the transcription of meetings, interviews, and lectures, as well as the intelligent generation of meeting minutes, achieving an accuracy rate of up to 97%. It is available across all platforms, with a continuous monthly subscription priced at 25 yuan.

2026-06-29

Adobe Podcast: The free AI tool that transforms smartphone recordings into studio-quality audio.

119

Adobe Podcast is a free AI-powered audio enhancement and podcast production tool from Adobe; its "Enhance Speech" feature can instantly elevate smartphone recordings to studio quality. It supports remote recording and text-based editing, and the free version is already highly capable.

2026-06-30

Proofig – a guardian of research integrity who uses AI to perform "CT scans" on academic papers, leaving no room for image fraud.

117

Proofig is an AI-powered academic image integrity detection platform designed specifically for scientific publications. It automatically identifies image tampering, duplication, plagiarism, and AI-generated images, and compares manuscripts against over 155 million publicly available images on PubMed. It is used by top publishers such as the Science family of journals for pre-submission image screening to maintain research integrity.

FigJam AI: A whiteboard partner that transforms brainstorming sessions from "boring" to "fast-paced".

116

FigJam AI is an AI suite for Figma whiteboards that can generate flowcharts, timelines, and meeting templates with a single click, automatically organizing notes and summarizing key points. It integrates with Claude AI for even smarter chart generation. Paid plans are available.

2026-06-29

Writer AI Content Detector—a completely free AI-powered "seven-dimensional scoring" tool, even GPTZero users use it as a backup.

114

Writer AI Content Checker is a completely free online tool for identifying whether text was generated by AI. It evaluates text across seven dimensions, providing a clear probability of whether it was written by a human or AI. No registration is required, there are no usage limits, and a convenient Chrome extension is also available. It's ideal for teachers, editors, and content creators who need to quickly and frequently screen content.

Flowith—a canvas-like creative space that liberates AI from chat windows, allowing you to complete all complex tasks with an infinite whiteboard.

112

Flowith is a canvas-based AI creation platform and agent system product that replaces traditional chat boxes with an infinite canvas. It features over 30 built-in models and offers innovative functionalities such as Neo Infinite Agents capable of performing tasks exceeding 1000 steps, a Knowledge Garden, and multi-agent collaboration. It has already secured tens of millions of dollars in funding. It is suitable for creators, researchers, and knowledge workers seeking deep creative work and automation of complex tasks.

StableLM—an open-source AI model from the same family as Stable Diffusion, capable of running with only 3B parameters and commercially viable.

112

StableLM is an open-source large language model series launched by Stability AI. It is trained on The Pile extended dataset, which contains 1.5 trillion tokens. It offers multiple parameter versions, including 1.6B, 3B, 7B, and 12B, and supports text generation and code writing. It is licensed under the CC BY-SA 4.0 open-source license, allowing free commercial use. Stable LM 2 12B outperforms Llama 2 70B in some benchmark tests. It is suitable for developers, researchers, and small and medium-sized enterprises for private deployment.

2026-07-02

SuperCLUE—A benchmark for Chinese large models: from foundational capabilities to AI agents, a single test reveals who is "swimming naked."

109

SuperCLUE is a comprehensive evaluation benchmark for general-purpose Chinese large models, released by the CLUE team. It provides authoritative, multi-dimensional capability assessments and rankings for Chinese large models by regularly publishing monthly and semi-annual reports based on three key benchmarks—open-domain multi-turn dialogue, closed-domain objective questions, and anonymous head-to-head battles—as well as dimensions such as mathematical reasoning, code generation, and AI agents.

2026-07-10

Tiangong AI — A platform featuring an agent model with a million-token context window and an intelligent canvas, enabling end-to-end delivery from demo to product.

104

Skywork is an AI agent platform under Kunlun Tech that integrates SkyClaw-v1.0 (an agent model supporting a million-token context window), the visual canvas Skywork Design, and Dynamic Workflows for parallel multi-agent scheduling. The platform covers the entire workflow—from research, design, and code generation to deployment—and the annual recurring revenue (ARR) from its AI-native business has surpassed $800 million.

2026-07-11

H2O.ai – From a pioneer in open-source machine learning to a guardian of "sovereign AI," redefining enterprise-level intelligent agents with a converged architecture.

100

H2O.ai is a leading enterprise AI platform that integrates predictive and generative AI, focusing on AI deployments utilizing private, protected data. Its flagship products, h2oGPTe and H2O Super Agent, enable the creation of secure, autonomous agents in on-premises, VPC, and air-gapped environments. These agents consistently rank at the forefront of accuracy in authoritative benchmarks such as GAIA and FutureX. The company is trusted by more than half of the Fortune 500 and organizations within the world's most highly regulated industries.

2026-07-10

prompts.chat—a "free and open-source" AI-powered suggestion community—is reshaping sharing rules with its protocols and code.

prompts.chat is a free, open-source, and self-hosted AI-powered prompting word social platform. The community focuses on sharing and discovering high-quality prompts, covering content from various fields including code, image generation, and writing. The platform emphasizes data privacy and community-driven development, providing a decentralized communication space for AI creators and users.

Cania: The world's first design specific basic model and intelligent design platform upgraded with

Cania is the world's first AI design model specifically designed for design, supporting conversational design. It can generate desired images, text, etc. using natural language and also has memory and learning capabilities!

2026-06-01

PubMedQA—an "AI benchmark" specifically designed for biomedical question answering; only by comprehending research papers can an AI truly pass the "Medical Turing Test."

PubMedQA is the first question-answering dataset requiring reasoning over biomedical research texts; it was released in 2019 by institutions including the University of Pittsburgh. The task involves answering "Yes," "No," or "Maybe" questions based on PubMed abstracts. Comprising 1,000 expert-annotated samples and 211,000 artificially generated ones, the dataset aims to evaluate the ability of AI models to comprehend and reason about complex medical literature. It is widely used to benchmark the performance of large language models in the medical domain.

2026-07-10

LLaMA—Meta's open-source AI model revolution, with performance comparable to GPT-4 and completely free for commercial use.

LLaMA is an open-source family of large language models launched by Meta, evolving from LLaMA 1 in February 2023 to multimodal LLaMA 4. The 70B parameter version boasts performance close to GPT-4, while the 405B version rivals GPT-4o. It is completely open-source and commercially viable. Enterprises can download and deploy it on their own servers without data sharing, meeting compliance requirements in industries such as finance and healthcare. With over 1.2 billion downloads globally, it is suitable for developers and enterprises requiring private deployment and deep customization.

2026-07-02

AnythingLLM – Full-stack AI applications and private knowledge bases, enabling you to build your own ChatGPT with any model.

AnythingLLM is a full-stack AI application that supports any commercial or open-source LLM. It features built-in RAG, AI agents, and no-code agent builders. It can run locally or be remotely hosted, intelligently converse with documentation, and requires no cumbersome setup. Supporting the MCP protocol, it offers Desktop and Docker deployment options, making it suitable for individuals and businesses to build private, fully functional AI assistants.

2026-07-14

iFlytek AI University – iFlytek's official AI learning and training platform, providing one-stop coverage from model fine-tuning to application development.

iFlytek AI University is an AI professional learning and training platform under iFlytek's open platform. Leveraging iFlytek's technological expertise in NLP, intelligent speech, and other fields, it offers a full-chain course from beginner to advanced model fine-tuning and agent development. The platform is deeply integrated with iFlytek's MaaS tools, supporting zero-code model customization and deployment. Upon completion, students receive official certification and it serves as a core hub for offline developer communities in multiple cities across China.

2026-07-14

Mooliv - AIGC Video Creation Platform with AI Editing, Cloud Editing and Massive Templates

Mooliv is an intelligent video creation platform based on AIGC technology, offering AI editing, cloud editing, and a vast library of video templates to help users quickly produce high-quality video content. No professional editing skills required—leveraging smart algorithms and rich assets, users can easily create short videos, marketing videos, and social media content, ideal for individual creators, marketing teams, and content professionals.

2026-07-17

There are 326 more tools not shown. Try more specific keywords.

AI News (10)

Meituan Open-Sources LongCat-2.0, a 1.6-Trillion-Parameter MoE Model; Full-Stack Execution on Domestic Computing Hardware Successfully Achieved

601

Meituan has officially open-sourced LongCat-2.0, a massive MoE model with 1.6 trillion parameters. This release achieves end-to-end training and inference deployment on a domestic computing cluster comprising 50,000 GPUs, thereby breaking reliance on overseas computing power. The announcement details LongCat-2.0's technical highlights, performance advantages, and industry value.

I used AI to draw avatars and create picture books for my child, and earned 10000 yuan in the third

236

Let me tell you a true story first. I have a friend named Xiaomin, a full-time mom with a two-year-old child. She has tried various side jobs before: being scammed for brushing orders, not being able to sell on WeChat, and not being able to write articles. At the end of last year, she told me,

OpenAI's GPT-5.6 series preview remains in internal testing, with proprietary inference chips set for mass production by year-end.

221

OpenAI is officially advancing the limited beta testing of its GPT-5.6 model series and has announced a partnership with Broadcom to co-develop "Jalapeño," a dedicated inference ASIC. The project targets tape-out within nine months and large-scale deployment by the end of 2026, aiming to slash inference costs by 50% and effectively completing the company's vertically integrated strategy combining models and chips.

Genesis World 1.0 is officially open source! Our self-developed simulation platform significantly shortens robot testing cycles.

216

Genesis AI, which gained fame for its robot frying tomatoes and eggs, has opened up its self-developed simulation platform, which greatly reduces the time required for robot evaluation and makes the simulation data closely match the real machine, thus helping to commercialize physical AI technology.

Anthropic's valuation has reached $965 billion, surpassing OpenAI for the first time.

211

The AI industry is undergoing a seismic shift as Anthropic’s post-investment valuation climbs to $965 billion, surpassing OpenAI for the first time. Its Claude Opus 4.8 model has reached the pinnacle of reasoning capabilities, while Claude Code captures over half the code-focused AI market with $6.3 billion in annual revenue; this analysis delves into the company's three core strengths: capital, technology, and commercialization.

The Rise of the "Vibe Coding" Wave: Ordinary People Can Independently Develop Full-Fledged Software Without Coding Skills

194

The "Vibe Coding" wave is here: ordinary people can now independently create websites and utility software using AI—simply by describing their requirements in natural language—without needing to master any actual code. This article explains the underlying principles, available tools, and real-world implementation examples.

Xu Li of SenseTime: AI billing will shift from tokens to tasks, establishing a new paradigm of inclusive AI.

162

SenseTime’s Xu Li: AI Billing to Shift from Tokens to Tasks, Establishing a New Paradigm for Inclusive AI At the main forum of the 2026 World Artificial Intelligence Conference, Xu Li, Chairman and CEO of SenseTime, delivered a keynote speech titled "Boundless Innovation and Bounded Safeguards: Inclusivity and Safety in AI Development." He put forward significant insights regarding the evolution of the AI industry, sparking widespread attention across the sector.

How to view the WALL-WM world model released by the independent variable robot? Can it really solve

119

To be honest, after reading the paper and demo, I immediately withdrew the paper on "Action Block Prediction" that I was planning to submit last week. Don't laugh, this is really painful for me. After six months of hard work, I thought about sewing and repairing another piece on the VLA (Visual

The independent variable robot releases the world's first "event level prediction" embodied

109

Breaking the limitations of traditional frame by frame learning, the way robots understand tasks has entered a new stage

36 billion US dollars! Apollo and Blackstone join forces to create AI computing power financing big

108

Global AI infrastructure welcomes unprecedented capital surge, with Broadcom and Google joining forces

Career Skills (109)

Multi-agent task allocation and scheduling capabilities

Clawdtributor is a built-in multi-agent scheduling core component of OpenClaw. It automatically breaks down complex tasks and creates batches of sub-agents for parallel execution. Supporting various load-balancing strategies and consolidating the verification of all sub-task results, it significantly boosts processing efficiency in scenarios such as bulk research, code auditing, and distributed data collection.

39.9w 2026-08-03

ClawSweeper: Automated Warehouse Work Order Cleanup and Inspection Capability

ClawSweeper is an intelligent DevOps agent skill designed for OpenClaw to automate issue management in GitHub projects. It automatically scans repository issues and pull requests (PRs) to identify invalid items—such as those that are stale, unreproducible, or already resolved—and uses AI to generate standardized closing comments for batch archiving, thereby reducing manual maintenance effort. It features built-in safety safeguards to prevent the accidental closure of issues marked with protected labels, linked to active PRs, or created by administrators. Additionally, it supports scheduled polling, manual trigger commands, issue status queries, and the automatic submission of fix PRs.

39.0w 2026-07-29

Security Bounty Hunter Skill

The Security Bounty Hunter skill is a specialized security skill within the ECC (Agent Harness Performance Optimization System), designed for security researchers and bounty hunters. It integrates multiple security testing tools and methodologies to help users efficiently discover vulnerabilities in web applications, APIs, mobile apps, and more. Core features include automated vulnerability scanning, intelligent payload generation, target reconnaissance and information gathering, and vulnerability report template generation. By leveraging advanced AI reasoning capabilities, this skill simulates attacker thinking, providing precise vulnerability verification steps and remediation suggestions. Suitable for penetration testing, bug bounty programs, security audits, and more, significantly improving security research efficiency.

100 2026-07-25

Healthcare PHI Compliance Skill

This skill focuses on Protected Health Information (PHI) compliance in the healthcare domain, helping developers and organizations ensure their AI agents and automation systems adhere to privacy regulations like HIPAA when handling sensitive medical data. It provides best practices for secure PHI handling, data masking methods, access control policies, and audit logging capabilities, suitable for healthcare, health tech, insurance, and bioinformatics scenarios to ensure data security and compliance.

100 2026-08-03

Perl Security Skill

The Perl Security Skill is a specialized skill within the ECC (Agent Harness Performance Optimization System), focusing on secure development and vulnerability protection for the Perl programming language. It provides core features such as Perl code security best practices, common vulnerability detection and remediation, input validation, output encoding, SQL injection protection, and XSS protection. Suitable for developers writing secure Perl scripts, security auditors, and teams looking to enhance the security of their Perl applications. By integrating this skill, you can automatically detect security risks in Perl code and receive remediation suggestions, ensuring the security of Perl applications in production environments.

100 2026-07-23

Maintain Hand Tools

This skill focuses on the maintenance and care of hand tools, covering operations such as cleaning, lubrication, adjustment, and repair. It is applicable to daily maintenance of various hand tools like wrenches, screwdrivers, pliers, and hammers. Through systematic procedures and checklists, it helps users extend tool lifespan, improve work efficiency, and ensure operational safety. Suitable for repair technicians, DIY enthusiasts, workshop managers, and more.

园林绿化

26 2026-07-18

HIPAA Compliance Skill

The HIPAA Compliance Skill is part of the ECC (Agent Harness Performance Optimization System), specifically designed to help AI agents and development tools meet the requirements of the Health Insurance Portability and Accountability Act (HIPAA). This skill provides guidelines, policies, and best practices for securely handling Protected Health Information (PHI), ensuring that AI applications developed in environments like Claude Code, Codex, Opencode, and Cursor comply with healthcare data privacy and security standards. Ideal for AI projects handling medical data, health tech applications, and healthcare IT systems.

0 2026-07-27

Property Care Agent Skill

This skill is designed for property management scenarios, helping users efficiently manage property assets, log chemical applications, understand various measurement units, and track seasonal maintenance tasks. It is suitable for professionals or enthusiasts in real estate management, gardening, agriculture, and related fields. Through structured logging and reminders, it ensures all operations are compliant and traceable, improving overall management efficiency. The skill includes built-in measurement conversions and seasonal calendars to reduce manual calculation and memory burden.

园林绿化

Post-Construction Clean Skill

The Post-Construction Clean Skill is an intelligent skill designed to guide users through post-construction cleaning tasks. Developed based on an open-source project on GitHub, it helps users efficiently and systematically handle dust, stains, and debris left after renovation. The skill provides detailed cleaning steps, tool lists, and precautions, covering a comprehensive cleaning process from walls, floors to furniture. Users can get personalized cleaning suggestions through natural language interaction, such as cleaning methods for different materials (e.g., tiles, wooden floors, glass) and how to deal with stubborn stains like paint spots and glue residue. Suitable for cleaning scenarios after new home renovation or old house remodeling, especially for ordinary users without professional cleaning experience, it saves time and effort while ensuring cleaning results meet habitable standards.

家政与清洁工

0 2026-07-29

Wardrobe Planner

Wardrobe Planner is a custom agent skill designed for AI agents, helping users efficiently manage their personal wardrobe. Core features include: clothing categorization and tagging, outfit suggestion generation, seasonal clothing recommendations, and shopping list planning. Users can use natural language commands to let the AI agent intelligently select outfit combinations based on weather, occasions, or personal preferences. Suitable for daily dressing, travel packing, wardrobe organization, and other scenarios, improving dressing efficiency and style consistency. This skill is based on an open-source project on GitHub and is easy to integrate into various AI workflows.

客房与清洁服务

Facade Documentation Skill

This skill creates before-after documentation and acceptance protocols for facade projects. Core features include: image comparison (before/after), condition checklists, water meter documentation, and PDF generation. It activates automatically with relevant requests and supports manual activation via specific commands. The skill is part of the FassadenFix ecosystem and is recommended to be used with fassadenfix-branding for CI guidelines. Suitable for professionals needing efficient management and recording of facade project documentation.

园林工人监督员

Landscaping Supervisor Skill

This skill is based on the domain-experts project on GitHub, aiming to transform human expert knowledge into AI agents. Specifically for the landscaping supervisor role, it provides a structured set of knowledge and instructions, enabling AI to simulate the decision-making, communication, and project management capabilities of a landscaping supervisor. Core functions include assisting with landscape design review, supervising construction progress and quality, managing teams and resources, and handling on-site issues. It is applicable to management scenarios in projects such as landscaping, municipal engineering, and private gardens, helping users improve supervision efficiency, reduce errors, and ensure projects meet design specifications and safety standards. With this skill, AI can act as a virtual supervisory assistant, offering professional advice and real-time feedback.

园林工人监督员

Fancy Yard Landscaper

Fancy Yard Landscaper is an intelligent skill package designed for AI agents, aimed at helping users plan and design beautiful yard landscapes. Based on the windags-skills project, this skill supports running in over 40 AI agents including Claude Code, Codex, Gemini CLI, and Cursor. It provides a complete set of skill instructions, scripts, and reference files to help AI agents understand fundamental yard design principles, plant selection, spatial layout, and aesthetic elements. Suitable for personal yard renovation, landscape design consultation, and automated design assistance scenarios. Users can install and activate this skill with simple commands, transforming AI agents into professional yard design assistants.

园林工人监督员

0 2026-07-27

Rehabilitation Analyzer

Rehabilitation Analyzer is an AI-powered skill for rehabilitation assessment and analysis, designed to assist therapists, doctors, or patients in quickly interpreting rehabilitation data, evaluating recovery progress, and providing personalized rehabilitation recommendations. It processes various rehabilitation inputs (such as range of motion, muscle strength test results, gait analysis data, etc.) and generates visual reports and improvement plans through intelligent algorithms. Suitable for clinical rehabilitation, home-based rehab training, sports injury recovery, and more, it significantly improves the efficiency and accuracy of rehabilitation evaluation.

物理治疗技能

0 2026-07-22

Managing Workers Compensation Rehabilitation

This skill is designed to manage and process workers' compensation rehabilitation workflows, assisting legal professionals, employers, or insurance companies in efficiently tracking injury cases, coordinating rehabilitation plans, calculating compensation amounts, and ensuring compliance with relevant laws and regulations. It integrates case recording, document generation, progress tracking, and other features, making it suitable for corporate HR departments, law firms, or insurance claims agencies. It simplifies complex administrative tasks in workers' compensation rehabilitation management, improving efficiency and reducing compliance risks.

物理治疗技能

0 2026-07-22

Skills in end-to-end release verification and CI monitoring

Release OpenClaw CI is a capability designed to execute, monitor, debug, and summarize the full release CI process for OpenClaw. Working in conjunction with `release-openclaw-maintainer` and `openclaw-testing`, it is utilized when release candidates require comprehensive validation, proof of installation/updates, real-time provider checks, or CI recovery. It handles the initiation, monitoring, troubleshooting, and evidence logging of full-matrix release validation but does not perform actual release operations—such as version bumps, tagging, `npm publish`, or GitHub releases—which require explicit operator approval.

39.9w 2026-07-12

OC-Path Test Fixture

This directory stores "real-world" input/output samples (fixtures) used by the unit tests for the oc-path extension; these samples serve to verify the accuracy and robustness of the path resolution engine when handling complex, cross-platform paths that include environment variables or special characters.

39.0w 2026-07-21

Node.js Inspect Debugger

Node.js Inspect Debugger is a powerful tool for debugging Node.js applications. It allows developers to connect to running Node.js processes via Chrome DevTools or other compatible debugging clients for breakpoint setting, variable inspection, call stack analysis, and performance profiling. Built on Node.js's built-in inspect protocol, it supports remote debugging, multi-process debugging, memory snapshots, and more, making it suitable for troubleshooting and performance optimization in development, testing, and production environments.

软件质量保证分析师与测试员

39.0w 2026-07-27

OpenClaw Test Performance Diagnostics

OpenClaw Test Performance is a specialized benchmarking, diagnostic, and optimization tool for OpenClaw test suites and plugin-suite runtimes. It helps developers quickly identify runtime hotspots, analyze CPU and RSS memory usage, detect heap growth trends, and uncover slow coverage paths. With one-click import into the Manus environment, developers can seamlessly integrate it into existing workflows for automated performance evaluation and tuning. This skill supports detailed metric collection, visual report generation, and intelligent optimization recommendations, making it ideal for continuous integration, performance regression testing, and code quality assurance scenarios.

软件质量保证分析师与测试员

38.6w 2026-07-24

Security Triage Skill

Security Triage is an Agent Skill from the OpenClaw project, designed to help AI assistants or users quickly perform initial classification and prioritization of security incidents, alerts, or potential threats. This skill uses a structured analysis process to identify the type of security issue (e.g., malware, network attack, data breach), assess its severity, and provide preliminary response recommendations. It is suitable for security operations personnel, incident response teams, and individuals or organizations looking to automate security classification workflows. Use cases include SOC (Security Operations Center) alert handling, penetration testing result classification, and preliminary vulnerability report analysis. This skill is based on an open-source project on GitHub and can be integrated into various AI platforms.

38.5w 2026-07-15

Techniques for batch-fixing high-certainty bugs within a limited scope

The "OpenClaw Small Bugfix Sweep" is a specialized workflow designed to batch-process issues and pull requests (PRs) in the OpenClaw repository by addressing minor, high-certainty bugs. Through in-depth code review, it targets only those bugs with clear root causes, unambiguous ownership, and minimal patch requirements. Upon fixing a bug, the system generates a "dirty diff" summary for manual review by Peter (the core maintainer); issues are submitted and closed only after his approval. Crucially, this process never involves pushing code, creating PRs, posting comments, closing issues, or applying tags during the review phase—all such actions are deferred until Peter grants explicit approval.

38.5w 2026-07-12

Alpha/Nightly Automated Release and Beta Triggering Capabilities

The OpenClaw Nightly release skill is an automated Alpha/Nightly release mechanism for OpenClaw/Tideclaw. It handles Alpha releases (triggered every 12 hours or manually), Beta release triggers, fixes on release branches, and forward-porting changes back to the `main` branch after a release. Working in tandem with the `release-private` skill, it utilizes the Tideclaw machine identity to execute releases on isolated branches, ensuring that a busy or broken state on the `main` branch does not impact the availability of Nightly builds.

38.5w 2026-07-12

Release a skill for automating changelog generation.

The OpenClaw Changelog Update is an automated utility designed to regenerate the OpenClaw release changelog from Git history. Prior to any beta or stable release (or re-release), it rewrites the relevant version section of `CHANGELOG.md` based on the complete history manifest, eliminating reliance on outdated draft notes. Used in conjunction with `release-openclaw-maintainer`, this utility manages changelog content, ordering, grouping, and attribution, serving as a mandatory step in the release process.

38.2w 2026-07-27

Maintainer Skills for Automated Review and Merging

OpenClaw PR Maintainer is a GitHub workflow automation tool designed for OpenClaw project maintainers to systematically review, triage, merge, and close issues and pull requests. It assists maintainers in efficiently managing external contributions while ensuring codebase quality and decision-making consistency through intelligent retrieval (prioritizing local archives), contributor identity analysis, deep code review (requiring an understanding of the full runtime context), and evidence-based merging criteria (mandating reproducible bug fixes).

38.2w 2026-07-12

Source-Audit-Driven Document Refactoring Skills

OpenClaw Refactor Docs is a skill designed for the source-audit-driven refactoring of existing OpenClaw documentation pages. Building upon the `openclaw-docs` skill—which handles style, page types, examples, and validation—it introduces a refactoring workflow that mandates the preservation and verification of all behavioral facts during rewriting, restructuring, splitting, or streamlining; this ensures that accurate descriptions of system behavior are not lost in the process. It requires source-based verification (against source code, tests, or schemas) for all behavior-sensitive information—such as configuration fields, commands, default values, and permission rules—rather than relying on inferences based on names or intent.