my-daily-news
Self-hosted GitHub AI project intelligence pipeline that automatically discovers trending AI apps, analyzes them via LLM, archives reports, and delivers daily summaries.
Ficha actualizada 22 sept 2026, 18:33 · upstream 22 sept 2026, 16:50 · release sin release detectada
Qué es my-daily-news
Descripción editorial construida a partir del repositorio, su documentación y sus señales de mantenimiento.
This tool automates the discovery and analysis of popular GitHub AI projects without requiring manual reading of every README or release. It combines GitHub Trending and Search API data with an LLM to filter for genuine AI applications, performs medium-depth analysis on each project, archives structured reports locally, and pushes concise daily highlights to a dashboard. The system supports pluggable models (Anthropic, OpenAI, Gemini, DeepSeek, Ollama), smart revisit logic that only re-studies projects with new releases or star growth, and flexible delivery channels including Feishu/Lark bots.
- Developers want to stay informed about emerging AI tools without manually scanning dozens of GitHub repositories daily
- Technical teams need structured intelligence on which AI applications are gaining traction in their stack
- Research analysts require a reliable, automated pipeline to track the evolution of AI agents and frameworks
- Product managers benefit from an objective view of what's trending vs. hype without relying solely on social media signals
- Dual-source ingestion combining GitHub Trending with Search API ensures comprehensive coverage of both new releases and search-driven projects
- LLM-based deep analysis reads READMEs, docs, and release notes to produce structured reports rather than generic summaries
- Smart revisit logic avoids redundant LLM calls by only re-studying projects that have new releases or significant star growth
- Project library maintains per-project metadata while daily digest focuses on the most impactful stories of each day
- Live dashboard auto-published to GitHub Pages with React + Tailwind styling for real-time monitoring
- Pluggable delivery via Feishu/Lark bots allows decoupling from third-party messaging platforms
- GitHub Trending API has rate limits that may require throttling during peak usage periods
- Search API availability depends on the GitHub search endpoint being operational and free-tier quotas are limited
- LLM analysis is not guaranteed to be perfect; false positives or negatives can occur depending on model quality and prompt design
- Projects with very low star counts may be filtered out by the LLM filter even if they contain AI content
- Feishu integration requires a custom app setup and approval process, which may delay delivery for new channels
Perfil de infraestructura
Estimación inicial derivada del repositorio y de los servicios detectados.
Despliegue detectado
Qué señales reales de instalación hemos encontrado.
Run as a Python script using `python -m src.pipeline` or via GitHub Actions workflow with proper secrets configured,For production use, consider adding rate limiting (e.g., `requests` library) to handle API quotas gracefully,Store LLM keys in environment variables and rotate them periodically; never commit sensitive credentials,The dashboard is built as a static React bundle that can be deployed independently or alongside the pipeline,For macOS, use launchd plist files with proper permissions (0755) to ensure the service runs on boot,Monitor GitHub API rate limits and adjust `collect.top_n` based on your organization's quota
Stack y dependencias
Componentes inferidos a partir de manifests, configuración y documentación.
Puede sustituir a
Relaciones detectadas por el análisis del repositorio y su categoría.
Experiencia real de la comunidad
Ayuda a contrastar la estimación automática con instalaciones reales.
Todavía no hay datos comunitarios para este proyecto.