Luo Jinbin
Community@DreamMacer
Luo Jinbin (DreamMacer) publishes 42 skills spanning software architecture governance, content production, financial analysis, Ascend NPU profiling, and agent skill authoring.
Agent Skills by Luo Jinbin
Showing 42 vetted skills indexed across 1 GitHub repositories.
DreamMacerskill-name-1
Analyzes delegated expert subtasks and returns evidence-backed conclusions with risks and next steps.
DreamMacercontent-enhancement
Enhances existing drafts with citations, opening rewrites, title optimization, and text polishing.
DreamMacercontent-drafting
Draft structured documents and chapters from outlines, notes, and interview records.
DreamMacercontent-review
Review manuscripts with paragraph-level diagnostics, structural analysis, and replaceable revision text.
DreamMaceroffice-document-workflow
Orchestrates parsing, conversion, editing, and redaction of PDF, Word, Excel, and PPT documents.
DreamMacerhealth-life-planning
Generates personalized diet, sleep, and stress management plans from user health profiles.
DreamMacerarch-review
Review architecture proposals against quality attribute checklists and produce risk-rated assessment reports.
DreamMacerarch-assess
Analyze existing codebases to map dependencies, draw as-is C4 diagrams, and assess architecture health.
DreamMacerarch-evolution
Plan architecture migrations, technical debt repayment, and governance workflows with phased roadmaps.
DreamMacerarch-engineering
Guides full-stack engineering practices covering layering, APIs, authentication, cloud infrastructure, and observability.
DreamMacerarch-design
Guides architecture selection and design using DDD, C4 diagrams, and ADR records.
DreamMacera-share-analysis
Analyzes A-share stocks, market sentiment, and sector trends using real-time market data.
DreamMacerfund-analysis
Analyzes mutual funds and builds portfolio allocation and dollar-cost averaging plans.
DreamMacerskill-creator-normal
Create, evaluate, and iteratively improve single-agent skills with benchmark analysis.
DreamMacerskill-gen-4-enterprise-doc
Generates installable agent skills from SOP documents, URLs, or stated intent.
DreamMaceropenJiuwen-DeepSearch
Generates multi-agent deep research reports in Markdown, HTML, and DOCX formats.
DreamMacerhuawei-cloud-maas-setup
Automates Huawei Cloud MaaS service purchase, API key creation, and model activation via browser scripts.
DreamMacergitcode-api
Query GitCode repositories, issues, pull requests, and files via the gitcode-api CLI.
DreamMacercross-channel-history-retrieval
Search cross-session chat history JSON files by channel, keywords, and time window.
DreamMaceradvanced-daily-report
Generate daily, weekly, and monthly work reports from Git, email, memory, and todo data.
DreamMacerswarmskill-creator
Creates, converts, and validates multi-role Swarm Skills with optional SwarmFlow orchestration scripts.
DreamMacerdynamic-memory-cli
Operate SQLite-backed agent memory with atomic Pending publication and Built index validation.
DreamMacerdeepep-to-cam-converter
Migrate DeepEP MoE dispatch and combine operators to CAM operators on Ascend NPU.
DreamMacerfinancial-document-parser
Extract structured data from invoices, receipts, and bank statements in PDF, image, and CSV formats.
Frequently Asked Questions About Luo Jinbin
FAQPage SchemaWhat tasks can I accomplish with DreamMacer's skill collection?âŒ
You can run architecture design, assessment, review, and evolution workflows (C4 diagrams, ADRs, strangler-pattern migration), draft and review long-form content, process Office documents, analyze A-share stocks and funds, profile Ascend MoE operators, build HarmonyOS apps, and author new single-agent or multi-agent skills.
Who is the target audience for these skills?âŒ
The collection targets software architects and full-stack engineers (arch-design/review/evolution/engineering), technical writers and editors (content-drafting/review/enhancement), financial analysts (a-share-analysis, fund-analysis), Ascend NPU kernel engineers, HarmonyOS developers, and agent platform operators building skills for JiuwenSwarm.
How do the architecture skills work together in practice?âŒ
They form a lifecycle: arch-design handles greenfield selection (microservices vs monolith, DDD, C4, ADR); arch-assess reverse-engineers existing systems; arch-review evaluates proposals against quality attributes and threat models; arch-evolution plans migrations and tech-debt governance; arch-engineering covers full-stack implementation concerns.
Are these skills open source and what do they cost?âŒ
The manifest specifies licensing only for project-maintainer, which is Apache-2.0. Other skills carry no explicit license field, so usage terms should be confirmed per repository. No pricing or paid-tier information is declared in the manifest.
What prerequisites or dependencies do these skills require?âŒ
Dependencies vary: huawei-cloud-maas-setup needs a Huawei Cloud account and bash; gitcode-api requires the gitcode CLI; cross-channel-history-retrieval needs mcp_exec_command and local session history files; a-share and fund analysis require the a_share_market data tool; Ascend trace analysis requires Chrome/Perfetto trace.json input.