supercli-site/xiaobai-kanban
- Source
- GitHub
- First trending
- Category
- Documents
- GitHub stars
- 580
- Main language
- HTML
This page introduces an external open-source repository. It is not an HDATF product.

What it does
xiaobai-kanban is a Chinese educational repository and documentation website about working with coding agents, not an operational kanban application. It targets beginners learning what agents are, how to express requirements, how to prepare tools and how to finish a first project. Learners provide natural-language instructions, background, goals, scope and acceptance criteria, supplemented by file locations, reference images and original error messages. A separately installed agent turns those instructions into file changes and command-line actions, while the first exercise produces an HTML page for the learner to inspect in a browser. The teaching method combines small changes, hands-on acceptance checks, explicit corrections, Git checkpoints and written handoffs rather than implementing a task-execution service. Reading requires a browser, while the documentation site's maintenance configuration uses Node.js, pnpm, VitePress and Cloudflare Pages. Extended exercises introduce authenticated GitHub, GitLab and Lark tools, plus optional Agent Mail with human account creation and authorization. The optional magpie chapter describes a local model gateway using provider accounts and API keys, with masking limited to routed requests and free usage dependent on provider quotas rather than account or subscription bypass.
License
MIT Permissive. Commercial use and changes are allowed if the copyright notice is kept.
More in this category
- google-research/paper-orchestra
PaperOrchestra is a multi-agent research manuscript writing framework, not an autonomous system that establishes experimental results. It takes idea and experiment Markdown files, conference guidelines and a LaTeX template, then produces manuscript files, figures when enabled and a final PDF. Its orchestrator links outline generation, literature retrieval, section writing and refinement, with literature and figure generation running concurrently in the plotting variant. Literature discovery uses Gemini with Google Search, followed by Semantic Scholar metadata retrieval and approximate title matching. The citation library relies on abstracts, truncates long abstracts and excludes records without abstracts, so a metadata match does not establish support from the full paper. Figure generation combines reference examples, plotting code or image generation and critique, but the plotting utility executes generated Python without an explicit sandbox boundary. The documented environment uses Python 3.11, model-provider credentials and LaTeX tooling, while refinement compiles drafts and uses model reviews and page images. The README's submission-ready claim is not a guarantee of scientific correctness, its generation dataset is absent from the repository and Google product support is explicitly disclaimed. - NYLLON-SOFTWARE/SignuDoc
SignuDoc is a document signing application for organizations that want to operate their own signing site and keep documents in their own hosting accounts. A sender uploads one PDF, identifies the recipients and their signing order, and places signature, date, text or selection fields. Signers open emailed links, consent to electronic records and complete their assigned fields in sequence, normally without creating accounts. The application tracks document status, deadlines and reminders, and can preserve document layouts and signer roles as reusable templates. Completion produces a signed PDF and audit certificates, with digital seals and file fingerprints for checking whether a retained copy has changed. The audit service links event records with SHA-256 hashes and checks sequence continuity, while the PDF finalization service checks the source file hash before applying field values. The production-oriented stack uses Next.js, Postgres, private Vercel Blob storage and Resend email, with Neon presented as a database option. Local development can instead store files on disk and capture outgoing messages without delivering them. It supports one PDF per envelope and sequential signing, while the README lists public API and webhooks as unavailable and does not promise that every document or jurisdiction accepts its signatures. - LibreTranslate/LibreTranslate
LibreTranslate is a self-hosted machine translation service for applications that need a translation API without depending on proprietary translation engines. It uses Argos Translate to process text on the server that the operator controls. Clients send text or a text list, source and target language codes, and a format to the translate endpoint, which returns translatedText. Repository tests cover single and batch requests and errors for missing text or unsupported languages. The Python package uses Flask, and the supplied Docker Compose configuration exposes the service on port 5000. Its initialization code downloads language packages when models need installation or updating, so offline operation requires preparing the required models first. A starting workflow prepares a self-hosted instance, checks available languages and submits a small translation request before sending documents through an application. Available language pairs depend on installed models, translation quality needs separate evaluation, and the project uses the AGPL-3.0 license. - wdcpclover/ai4paper
AI4Paper is a research reading and writing service for people who otherwise move between literature search, PDF reading, notes and review drafting. Its browser workspace and Zotero plugin accept papers, bibliographic files or DOI lists and provide library organization, translation, explanations, source-linked notes and writing assistance. The README describes expanding Chinese search terms into English queries, searching multiple sources and using citation networks to prepare review material. Official tutorials show a selected-paper review workflow whose input depth can be titles, abstracts or reading notes, so an output should not automatically be treated as full-text synthesis. Users need an account, and the plugin route requires a compatible Zotero installation; some functions require membership and consume AI quota. The product advertises cloud model options and a local Ollama option, but this does not establish that every search, translation and account function works offline. The repository ends with an all-rights-reserved notice and does not establish an open-source reuse license for the complete service. Its own review tutorial says it did not actually generate the review during that walkthrough, and citation accuracy, translation quality and the advertised literature corpus were not independently verified here. - FluentRead/FluentRead
FluentRead is a browser extension for readers who need translations beside the original rather than a separate translated page. It accepts webpage passages, selected text and supported documents, and returns bilingual reading views with optional AI explanations, image text recognition and video subtitle tools. Its website adaptation catalog combines shared content profiles with site-specific rules to select reading areas and protect code, forms and navigation. Translation text goes to the chosen service, which can be a free service, a configured cloud API, an AI provider or a supported local model. Users select languages and a provider after adding the extension, while AI explanations require a working AI configuration and provider charges remain separate. The document reader supports PDF, ePub and DOCX among other formats, with file-size limits, editable translations and exports. Recognition errors, technical terms, changing website structure and differences between extension and userscript features still require review. Current public releases do not offer ChatGPT subscription sign-in, and cloud translation can send selected content and permitted context outside the browser.
Only repositories in the ranked Trendshift lists are included, and the lists are used only to find candidates. We do not copy their ranks. Descriptions, licenses and star counts come from each GitHub repository. The notes are our own reading. We have not tested these projects, and a place on a trending list does not prove quality.