awesome-repositories.com
Blog
MCP
awesome-repositories.com

Discover the best open-source repositories with AI-powered search.

ExploreCurated searchesOpen-source alternativesSelf-hosted softwareBlogSitemap
ProjectMCP serverAboutHow we rankPress
LegalPrivacyTerms
© 2026 Bringes Technology SRL·VAT RO45896025·hello@awesome-repositories.com
imraywang avatar

imraywang/wewrite

0
View on GitHub↗

Wewrite

WeWrite is a modular suite of agent skills for writing WeChat Official Account articles. It takes a piece from topic selection through sourced drafting, editorial review, themed formatting and delivery to the platform's draft box, and it can run the whole sequence from a single instruction. Every factual claim is linked to a saved source in a per-article ledger, facts are separated from opinions, and drafts with invented experiences or unsupported conclusions are blocked.

Several things set it apart. Drafts are scored on a one-to-five rubric covering accuracy, viewpoint, usefulness, voice fit and readability, then rewritten and re-reviewed until they pass. Eleven mechanical checks flag clichés, fragmented sentences and repetitive rhythm. A style system learns an author's voice from manual edits, sample articles and reading data, and a change becomes a stable rule only after it recurs or is confirmed. Seven selectable personas are also available. Publishing is gated: nothing is sent to the draft box unless the user has explicitly allowed it, and otherwise a local HTML file is produced.

Formatting converts Markdown to inline-styled HTML using eighteen themes with dark mode, rich layout blocks such as timelines and callouts, and checks that catch constructs the platform would filter out. Custom themes can be extracted from a published article's address. Cover and in-article images can be generated with AI after a count and cost check, without altering the original text.

Other areas include topic discovery from real-time hot searches with ten ranked, deduplicated candidates, search-optimized titles and tags, web research, and importing existing articles as Markdown. Analytics pulled from published articles feed future topic choices, and articles can be repurposed into image-text or short-video versions. Each job lives in its own resumable run folder, and each skill can run alone. A command-line interface works without an agent for previewing, publishing, scoring, validating and diagnosing.

AI search

Explore more awesome repositories

Describe what you need in plain English — the AI ranks thousands of curated open-source projects by relevance.

Start searching with AI

Features

  • WeChat Article Typesetting - Runs a WeChat official account article from research through sourced drafting, review and formatting from a single instruction.
  • WeChat Article Writing Agent Skills - Provides modular agent skills that carry WeChat Official Account articles from topic selection through sourced drafting, editorial review, themed formatting, and draft-box publishing.
  • Claim Provenance Trackers - Separates facts, inferences, opinions and experience, links each claim to a saved source, and drops unsupported specifics from drafts.
  • Evidence-Backed Deliverable Drafting - Checks drafts for accuracy, viewpoint, usefulness, voice and readability, blocks invented experiences and unsupported claims, and rewrites until approved.
  • Evidence-Gated Claim Ledgers - Links every factual claim to a saved source in a per-article ledger and strips unsupported specifics before the draft moves on.
  • AI Article Generators - Turns a single prompt into a sourced, reviewed article through a multi-step pipeline with isolated, resumable task folders recording each step.
  • Article Theme Libraries - Turns Markdown into themed styled HTML with cover images and sends the finished article to the draft box once publishing is allowed.
  • Long-Form Article Draft Uploaders - Runs topic selection, sourced writing and editorial review in sequence from a single instruction, with optional pauses to confirm topic or outline.
  • Inline-Styled Output - Converts Markdown to HTML with all styling inlined from selectable themes, plus fixes for constructs the platform filters out.
  • Themed Inline-Style Renderers - Converts Markdown into inline-styled HTML using one of eighteen themes, with dark-mode support and automatic fixes for platform rendering problems.
  • Rubric-Gated Draft Reviews - Scores drafts on accuracy, viewpoint, usefulness, voice and readability, then rewrites and re-reviews until a pass verdict is reached.
  • Sourced Article Drafting - Builds a brief with reader, core judgment and boundaries, records claims against saved sources, and drafts from one of seven structural frameworks.
  • Style Learning - Learns style rules from manual edits, sample articles and reading data, making rules stable only after recurrence or confirmation.
  • Composite Skill Workflows - Splits the workflow into independent skills that run alone or chain together, filling in missing prerequisites automatically.
  • Evidence-Linked Research Pipelines - Searches the web to verify data, quotes and cases, then saves a per-article ledger tying each claim to its supporting evidence.
  • Isolated Run Workspaces - Gives each job an isolated folder holding its brief, source ledger, permissions and step statuses so parallel runs never collide.
  • AI Illustration Generators - Generates a cover and needed in-article images with AI after a count and cost check, saving an illustrated copy without changing the original text.
  • Trending Search Retrievers - Fetches real-time hot-search lists from several platforms to surface current subjects worth writing about.
  • Content Source Traceability - Saves the brief, claims list with source status, draft, final article and review report so every step stays traceable.
  • Critical Flaw Rejection Gates - Rejects drafts outright when they contain invented author experiences, unsourced core facts, conclusions beyond the evidence, or no clear takeaway.
  • Prose Style Linting - Runs eleven automated pattern checks that flag clichés, fragmented sentences and repetitive rhythm as mechanical risk hints.
  • Draft Review and Rewrite Loops - Judges drafts on accuracy, stance, usefulness, voice, and readability, rewrites failing ones, and re-reviews them to produce a final article with an edit report.
  • Markdown to HTML Renderers - Renders Markdown into inline-styled HTML using eighteen themes with dark mode and fixes for rendering problems on the publishing platform.
  • Prose Style Linters - Runs eleven automated checks that locate clichés, fragmented sentences, and repetitive rhythm, giving mechanical risk hints without judging who wrote the text.
  • Publishing Permission Gates - Records an explicit allow or deny publishing decision on each run, set only when the user requests or withdraws publication, and blocks delivery otherwise.
  • Rubric-Based Draft Scoring - Rates each draft from one to five on accuracy, viewpoint, usefulness, voice fit, and readability, passing only when average and individual scores clear thresholds.
  • Topic Idea Rankers - Produces ten candidate topics scored on three dimensions from trending searches and demand signals, deduplicated against history before drafting.
  • Writing Voice Consistency - Keeps an author's voice steady across posts by learning from edits and sample articles and offering selectable writing personas.
  • Trending Topic Detectors - Pulls real-time hot searches, estimates demand and ranks ten deduplicated candidate topics before drafting begins.
  • Stateful Pipeline Executions - Creates an isolated run folder holding article path, source ledger and permissions, merges module output, and records every step's status.
  • Action Permission Gates - Blocks publishing unless an explicit allow decision is recorded on the run, falling back to a local HTML file otherwise.
  • Resumable Task Frameworks - Stores each job in its own task folder with saved progress so runs proceed in parallel and failed ones resume where they stopped.
  • Publishing Platform HTML Compliance Checks - Checks generated HTML for constructs the target publishing editor would filter out and blocks them before preview or publishing.
3,668 stars·576 forks·Python·MIT·9 views

Star history

Star history chart for imraywang/wewriteStar history chart for imraywang/wewrite

How this analysis was created: This summary and feature list are AI-generated from collected project material and can contain mistakes. Stars, license and language are imported from GitHub. Inclusion does not mean that we have tested or audited this project. Check the source documentation for any feature you depend on. Learn more on our About page.

Frequently asked questions

What does imraywang/wewrite do?

WeWrite is a modular suite of agent skills for writing WeChat Official Account articles. It takes a piece from topic selection through sourced drafting, editorial review, themed formatting and delivery to the platform's draft box, and it can run the whole sequence from a single instruction. Every factual claim is linked to a saved source in a per-article ledger, facts are separated from opinions, and drafts with invented experiences or unsupported conclusions are blocked.

What are the main features of imraywang/wewrite?

The main features of imraywang/wewrite are: WeChat Article Typesetting, WeChat Article Writing Agent Skills, Claim Provenance Trackers, Evidence-Backed Deliverable Drafting, Evidence-Gated Claim Ledgers, AI Article Generators, Article Theme Libraries, Long-Form Article Draft Uploaders.

Which projects share features with imraywang/wewrite?

Projects with overlapping indexed features include: isjiamu/gzh-design-skill — This is an AI agent skill that turns written drafts into paste-ready HTML for WeChat Official Account articles. It… trycompai/crm — This is an open-source, self-hosted customer relationship manager built around autonomous AI agents. A durable… mcncarl/yichen-skills — This repository is a collection of reusable skills that extend AI coding agents. The skills cover cross-platform web… vikiboss/60s — This project is a multi-purpose REST API utility collection and developer suite. It serves as a centralized service… openaispace/ai-trend-publish — This project is an automated content automation pipeline and AI article generator. It uses large language models to… yaojingang/geoflow — GEOFlow is a self-hosted content management platform for generating, quality-checking, publishing and maintaining…

Projects sharing features with Wewrite

These projects share indexed features with Wewrite. Shared tags can include platform or build tooling; verify the primary use case before treating a result as a replacement.
  • isjiamu/gzh-design-skillisjiamu avatar

    isjiamu/gzh-design-skill

    4,287View on GitHub↗

    This is an AI agent skill that turns written drafts into paste-ready HTML for WeChat Official Account articles. It converts Markdown into inline-styled markup, so headings, quotes, lists, images, code blocks, and animated GIFs keep their layout when pasted into the platform's rich-text editor. Word and PDF drafts are also accepted: their structure is normalized first, then a theme matched to the document's topic is applied. Styling comes from a library of six curated themes. Each is a full component library with design variables, reusable HTML blocks, and rules that map Markdown elements to s

    HTMLagent-skillai-agentclaude-code
    View on GitHub↗4,287
  • mcncarl/yichen-skillsmcncarl avatar

    mcncarl/yichen-skills

    5,204View on GitHub↗

    This repository is a collection of reusable skills that extend AI coding agents. The skills cover cross-platform web research, local WeChat chat data analysis, persistent agent memory, social bookmark archiving, two-agent coding collaboration, and publishing articles to X as drafts. Many of them are built around guarded workflows: search results stay unverified candidates until checked against original sources, login-based fallbacks stay off until the user approves them for a task, and external agents run in sandboxed sessions with no copied credentials. The collaboration skills pair a ChatGP

    Python
    View on GitHub↗5,204
  • trycompai/crmtrycompai avatar

    trycompai/crm

    11,892View on GitHub↗

    This is an open-source, self-hosted customer relationship manager built around autonomous AI agents. A durable background agent with its own work queue researches contacts, companies and deals, decides what to examine next, schedules follow-up rechecks with a stated reason shown to the sales rep, and stays within a research budget while no one is present. Each contact, company and deal also has a chat tab where you can talk to the agent, see its steps, discarded leads and questions, and keep the conversation across page reloads. What sets it apart is how it treats what the agent finds. Facts

    TypeScript
    View on GitHub↗11,892
  • vikiboss/60svikiboss avatar

    vikiboss/60s

    4,780View on GitHub↗

    This project is a multi-purpose REST API utility collection and developer suite. It serves as a centralized service for real-time information aggregation, data transformation, and a wide array of programmatic tools. The service distinguishes itself by providing a broad range of specialized content delivery endpoints, from curated daily summaries and global trending rankings to randomized entertainment content like jokes and quotes. It also functions as a real-time aggregator for environmental and network data, including weather forecasts, currency exchange rates, and public IP lookups. The c

    TypeScript60sapibilibili
    View on GitHub↗4,780
Compare all 30 related projects→