Skill Claude CodeCodex
A matching workflow that compares an AI project request with 240 technology partners in central and southern China. It explains suitable partners and produces a report in HTML, a format that can be opened in a web browser.
Skill Claude CodeCodex
A research workflow for tracking how electricity, grid, oil, gas, and renewable-energy companies use large language models, AI systems that generate and analyze text.
Skill Claude CodeCodex
An image-analysis assistant that examines a supplied picture and turns its visual details into prompts for ERNIE-Image. A prompt is the written instruction used to generate an image.
Skill Claude CodeCodex
Generate images with ERNIE Image through a configurable service endpoint. Use this skill when the user wants Chinese-first text-to-image generation, multilingual text rendering, long-text or multi-region text in images, or when the service base URL, auth token env name, model path, and ERNIE-Image weight tar path must…
Skill Claude CodeCodex
This skill should be used when the user wants to monitor, search, or analyze social media and web content about ERNIE-Image across Chinese platforms including Zhihu, Xiaohongshu, WeChat Official Accounts, and Baidu Search.
Skill Claude CodeCodex
A monitoring and public-opinion analysis workflow for ERNIE, Baidu’s large-language-model product family, and related competitors. It collects posts, articles, videos, news, and technical discussions from multiple platforms, then analyses sentiment, influencers, risks, and competitor activity.
Skill Claude CodeCodex
A data-collection and analysis workflow for finding statements by experts in government, think tanks, and media. It evaluates their relevance and authority in a chosen field and ranks them for priority.
Skill Claude CodeCodex
Build and maintain an end-to-end expert-operations content collection and strategy synthesis pipeline. Use this skill when the user clearly wants the full workflow, or a substantial subset of it: building or maintaining a Feishu Base for expert-ops sources, collecting multi-source expert-ops content, scoring and…
Skill Claude CodeCodex
Use when building or maintaining a Feishu Bitable workflow that monitors public statements from government, ministries, think tanks, universities, and authoritative media, then writes clean expert and statement records with clickable links, source tags, priority labels, and expert aggregation.
Skill Claude CodeCodex
Build premium frontend-based technical diagram systems for proposals, whitepapers, presales demos, interactive architecture explainers, and exportable presentation assets. Always use this skill when the user wants to turn technical content into polished diagrams or a live demo page, especially if they mention solution…
Skill Claude CodeCodex
A GitHub tracking workflow for reviewing recent pull requests, which are proposed code changes, in the vLLM and FastDeploy projects. It produces a short comparison of their technical progress.
Skill Claude CodeCodex
Use this skill when the task is to discover, analyze, classify, or structure Hubei or Wuhan enterprise leads for AI ecosystem cooperation. Apply it for enterprise clue collection, AI demand judgment, partner prioritization, lead-table normalization, or SOP-driven screening tied to Baidu PaddlePaddle, regional AI…
Skill Claude CodeCodex
A research workflow for investigating international leads such as AI agents, companies, platforms, and tools. It gathers information from their websites, GitHub, and social media, then examines possible cooperation with PaddleOCR or ERNIE.
Skill Claude CodeCodex
A step-by-step workflow for turning a Chinese intellectual-property novel into a short AI-generated video series. It covers story analysis, scripts, shot lists, character designs, images, voice-over, and video assembly, with user approval at each stage.
Skill Claude CodeCodex
A recurring research workflow for tracking the large-language-model data market, including training data, data-labeling recruitment, and expert recruitment. It collects recent information from Chinese social and media platforms, organises it in Excel, and produces market insight reports.
Skill Claude CodeCodex
A LaTeX document workflow for producing customer-facing PDF technical proposals about enterprise large-language-model solutions. LaTeX is a document system used for consistent, traceable formatting of technical and formal documents.
Skill Claude CodeCodex
A briefing generator for important meetings involving senior leaders. It combines information about the meeting, both sides’ goals, the organisation or person being met, the regional context, and possible cooperation.
Skill Claude CodeCodex
A writing assistant for creating 300–500-character Chinese micro-fiction from the emotions, setting, people, objects, actions, and time supplied by the user. It supports healing, sharp, and letting-go tones.
Skill Claude CodeCodex
A research workflow for investigating government bodies, companies, research institutions, and other organizations. It collects public evidence, checks links, and turns approved facts into a meeting-preparation report.
Skill Claude CodeCodex
A process for turning Chinese technical material into English Twitter posts for PaddlePaddle. It covers source reading, content classification, writing, image handling, preview, and review.
Skill Claude CodeCodex
Expert guide for PaddleOCR, PaddleX, PP-OCRv5, PP-StructureV3, PP-ChatOCRv4, table recognition, and PaddleOCR-VL. Use this skill when the user asks about OCR, PDF parsing, Markdown/JSON/XLSX export, model or pipeline selection, service deployment, hardware compatibility, PaddleX configuration, troubleshooting…
Skill Claude CodeCodex
A guide for choosing how to run PaddleOCR's PP-OCR text-recognition models on phones, embedded Linux devices, browsers, or cross-platform apps. PaddleOCR is software that reads text from images.
Skill Claude CodeCodex
A Chinese-language writing guide for messages that propose partnerships, such as outreach emails, follow-ups, joint events, and co-created content.
Skill Claude CodeCodex
A guided process for handing over a project or job, turning the departing person's knowledge into documents and searchable agent memory.