TrendingAgent

GitHub Trending Signal Intelligence

AI-powered daily intelligence for discovering breakout open-source projects, decoding developer momentum, and mapping emerging technology trends.

←
→LatestObserved2026-06-20

Leaderboard

Top 15 repositories

Ranked by GitHub Trending today so you can quickly scan heat, language, and new stars.

15 repos
#1DeusData/codebase-memory-mcpA high-performance code intelligence MCP server designed for AI coding agents, enabling millisecond-level codebase indexing and querying via knowledge graphs.+1,058#2google-research/timesfmA pretrained time-series foundation model developed by Google Research for time-series forecasting.+1,510#3palmier-io/palmier-proAn open-source, Swift-native video editor for macOS designed for AI-integrated workflows.+756#4koala73/worldmonitorAn AI-powered real-time global intelligence dashboard that aggregates news and monitors geopolitics and infrastructure.+156#5aishwaryanr/awesome-generative-ai-guideA one stop repository for generative AI research updates, interview resources, notebooks and much more!+107#6BuilderIO/agent-nativeA framework for building agent-native applications.+147#7chopratejas/headroomA local-first library that compresses context for AI agents and LLM applications to drastically reduce token usage.+4,005#8calesthio/OpenMontageAn open-source agentic video production system that turns AI coding assistants into full video studios.+156#9zai-org/GLM-5The GLM-5 series is a new generation of open-source large language models focused on long-horizon tasks and agentic engineering.+480#10withastro/flueA TypeScript sandbox agent framework for building autonomous AI agents and powerful workflows.+309#11n0-computer/irohA modular networking stack in Rust that enables dialing by public key instead of IP addresses.+302#12obra/superpowersAn agentic skills framework providing a systematic software development methodology for coding agents.+1,110#13penpot/penpotAn open-source design platform for design and code collaboration, featuring self-hosting and real-time collaboration.+85#14Kong/insomniaAn open-source, cross-platform API client for GraphQL, REST, WebSockets, SSE, and gRPC.+292#15Lightricks/LTX-2Official Python inference and LoRA fine-tuning package for the first DiT-based synchronized audio-video generative foundation model.+196

Languages

Language distribution

Ranking

New Stars Today Top 10

Ranking

Total Stars Top 10

Daily Brief

Today's brief

Fetched 15 GitHub Trending repositories today. The LLM daily brief is currently unavailable, so this is a rule-based summary from descriptions, languages, and star data. Main languages include: C, Python, Swift, TypeScript, HTML.

Repositories

Trending repositories

#1

DeusData/codebase-memory-mcp

A high-performance code intelligence MCP server designed for AI coding agents, enabling millisecond-level codebase indexing and querying via knowledge graphs.

8,272

Stars

629

Forks

+1,058

Today

README summary

This project is a high-performance code intelligence engine for AI coding agents. It indexes codebases (like the Linux kernel) into a persistent knowledge graph in minutes and responds to structural queries in sub-millisecond time, supporting 158 languages. Its core advantages are extreme indexing speed, a single static binary with zero dependencies, and automatic integration with 11 major coding agents. It's relevant for AI developers and tool integrators who need deep structural understanding of codebases.

Why it matters

Ranked #1 on GitHub Trending, the project has gained significant attention today (1058 stars). It introduces an innovative knowledge graph approach that substantially improves the efficiency and capability of AI agents interacting with codebases, representing an important technical direction in AI-assisted programming tools.

C

#2

google-research/timesfm

A pretrained time-series foundation model developed by Google Research for time-series forecasting.

24,099

Stars

2,274

Forks

+1,510

Today

README summary

TimesFM is a time-series foundation model from Google Research designed for forecasting. Its core capabilities include supporting context lengths up to 16k, continuous quantile forecasting, and covariate support via XReg. It is relevant for developers and data scientists seeking powerful and scalable time-series prediction.

Why it matters

Developed by Google Research and integrated into their cloud products (like BigQuery ML), indicating enterprise-level validation. The latest model significantly improves context length and forecasting capabilities while reducing parameter count, and is actively maintained with support for fine-tuning and community fixes, suggesting strong technical momentum.

Python

#3

palmier-io/palmier-pro

An open-source, Swift-native video editor for macOS designed for AI-integrated workflows.

1,955

Stars

192

Forks

+756

Today

README summary

Palmier Pro is an open-source video editor built from scratch in Swift for macOS on Apple Silicon. Its core capabilities include integrated state-of-the-art generative AI models and seamless integration with external AI agents like Claude, Codex, and Cursor via the MCP protocol for collaborative editing. It is aimed at developers and content creators looking to explore AI-assisted workflows within a professional video editing environment.

Why it matters

The project is ranked highly on GitHub Trending with a significant single-day star increase (756), indicating strong community interest in its 'AI-native video editor' positioning. Its innovative use of the MCP protocol to connect the editor with various AI agents represents a new direction in AI tool integration.

Swift

#4

koala73/worldmonitor

An AI-powered real-time global intelligence dashboard that aggregates news and monitors geopolitics and infrastructure.

57,278

Stars

9,141

Forks

+156

Today

README summary

This project is a real-time global intelligence dashboard that uses AI to aggregate news, monitor geopolitical and infrastructure changes, and provide a unified situational awareness interface. Its core capabilities include over 500 curated news feeds across 15 categories, dual map engines (3D globe and flat map), cross-stream correlation analysis, a Country Instability Index, a financial radar, and support for running local AI models. It is suitable for developers and tech observers who require real-time global situational awareness, geopolitical analysis, or financial and energy market monitoring.

Why it matters

The project is comprehensive, integrating multi-dimensional data from news, geopolitics, and finance, offering AI analysis and various application forms. Its modern tech stack and support for local deployment make it highly practical and noteworthy.

TypeScript

#5

aishwaryanr/awesome-generative-ai-guide

A one stop repository for generative AI research updates, interview resources, notebooks and much more!

27,642

Stars

5,739

Forks

+107

Today

README summary

The README was fetched, but the LLM summary is currently unavailable. Open the repository for details.

Why it matters

It gained 107 stars today and ranks #5. A one stop repository for generative AI research updates, interview resources, notebooks and much more!

HTML

#6

BuilderIO/agent-native

A framework for building agent-native applications.

1,060

Stars

119

Forks

+147

Today

README summary

The README was fetched, but the LLM summary is currently unavailable. Open the repository for details.

Why it matters

It gained 147 stars today and ranks #6. A framework for building agent-native applications.

TypeScript

#7

chopratejas/headroom

A local-first library that compresses context for AI agents and LLM applications to drastically reduce token usage.

38,799

Stars

2,648

Forks

+4,005

Today

README summary

Headroom is a context compression layer designed to compress tool outputs, logs, files, and RAG chunks before they reach the LLM, claiming to reduce token usage by 60-95% while preserving answer quality. It offers multiple integration modes including a library, proxy, and MCP server, with six compression algorithms and cross-agent memory support. This project is for developers and tech observers focused on optimizing AI agent development costs, LLM application performance, and token consumption control.

Why it matters

The project directly addresses the pain point of high token consumption and cost in LLM applications, offering multiple practical integration modes and demonstrating significant token savings and benchmark results on real workloads, giving it clear utility and relevance.

Python

#8

calesthio/OpenMontage

An open-source agentic video production system that turns AI coding assistants into full video studios.

6,314

Stars

1,084

Forks

+156

Today

README summary

OpenMontage is an open-source agentic video production system that transforms your AI coding assistant (like Claude Code or Cursor) into a full video production studio. Featuring 12 pipelines and 52 tools, it can handle research, scripting, asset generation, editing, and final composition through natural language descriptions. The project can create both image-based animations and videos using real motion clips sourced from free stock footage and open archives. It is aimed at developers and tech enthusiasts who want to leverage AI for rapid and low-cost video creation.

Why it matters

The project integrates AI coding assistants with a full video production workflow, demonstrating the powerful potential of agents in creative pipelines and lowering the barrier and cost of video creation.

Python

#9

zai-org/GLM-5

The GLM-5 series is a new generation of open-source large language models focused on long-horizon tasks and agentic engineering.

4,592

Stars

463

Forks

+480

Today

README summary

The GLM-5 series (including 5.2, 5.1, and 5) are large language models designed for complex systems engineering and long-horizon agentic tasks. Their core strengths include strong coding, reasoning, and long-context handling (up to 1M tokens), achieving state-of-the-art performance among open-source models on several benchmarks. The series is targeted at developers and researchers interested in AI agents, long-horizon task automation, and open-source large model deployment.

Why it matters

Released by zai-org, this is a flagship open-source model series focused on solving long-horizon, complex engineering tasks. GLM-5.2 demonstrates strong performance on multiple coding and agentic benchmarks and provides detailed local deployment guides, offering high reference value for the community interested in cutting-edge open-source AI models and agent development.

Unknown

#10

withastro/flue

A TypeScript sandbox agent framework for building autonomous AI agents and powerful workflows.

5,847

Stars

324

Forks

+309

Today

README summary

Flue is a TypeScript agent harness framework designed to provide models with the context and environment (like sessions, tools, skills, instructions, and a secure sandbox) needed for truly autonomous work. It offers features for building agents, workflows, subagents, and connecting to MCP servers, with deployment support on various platforms like Node.js and Cloudflare Workers. Developers interested in building next-generation autonomous AI agents and automated workflows should pay attention to it.

Why it matters

The project is experiencing rapid recent growth (309 stars today), focusing on the hot topic of AI agent architecture. It provides a comprehensive TypeScript toolkit for building, deploying, and monitoring autonomous agents, showing an evolution from raw API calls to a more mature framework.

TypeScript

#11

n0-computer/iroh

A modular networking stack in Rust that enables dialing by public key instead of IP addresses.

10,254

Stars

466

Forks

+302

Today

README summary

Iroh is a modular networking library providing an API for dialing by public key, automatically finding and maintaining the fastest connection to a target device, even behind NATs. Its core capabilities include built-in hole-punching, QUIC-based encrypted and multiplexed connections, and composable higher-level protocols like content-addressed blob transfer and gossip publish-subscribe networks. It is suitable for Rust developers building peer-to-peer networks, decentralized applications, or those prioritizing connection reliability.

Why it matters

The project addresses the instability of traditional IP networking through public key addressing and intelligent connection management. Its modular design (iroh-blobs, iroh-gossip) and use of QUIC represent a significant direction for modern P2P networking stacks. The notable recent star growth (302/day) indicates high developer community interest.

Rust

#12

obra/superpowers

An agentic skills framework providing a systematic software development methodology for coding agents.

233,389

Stars

20,724

Forks

+1,110

Today

README summary

Superpowers is a complete software development methodology for coding agents, built on composable skills and instructions that guide them through structured workflows, from brainstorming and planning to subagent-driven, test-first implementation. Its core capabilities include modular skills like brainstorming, plan writing, subagent-driven development, and enforced test-driven development. The project targets developers using various coding agents (e.g., Claude Code, Cursor, Codex) seeking to add structure and autonomy to their development process.

Why it matters

The project's high rank and significant daily star growth on GitHub Trending suggest it addresses a common pain point for developers using AI coding assistants: how to systematically guide and constrain agent behavior. Its methodology, emphasizing test-driven development and plan-first workflows, offers practical value for teams focused on code quality and efficiency.

Shell

#13

penpot/penpot

An open-source design platform for design and code collaboration, featuring self-hosting and real-time collaboration.

50,609

Stars

3,268

Forks

+85

Today

README summary

Penpot is an open-source design platform for teams building digital products at scale. Its core strength is giving teams full ownership of their design infrastructure through self-hosting, real-time collaboration, and open standards (SVG, CSS, HTML). Key capabilities include native Design Tokens, CSS Grid/Flex layouts, a plugin system, and a powerful API to create a full-stack platform connecting design, code, and AI workflows.

Why it matters

The repository is ranked high on GitHub Trending with 85 stars gained today, indicating rising interest. As an open-source design tool, it emphasizes self-hosting, open standards, and design-code collaboration, making it a key option for technical teams that value data sovereignty and customizable workflows.

Clojure

#14

Kong/insomnia

An open-source, cross-platform API client for GraphQL, REST, WebSockets, SSE, and gRPC.

38,987

Stars

2,300

Forks

+292

Today

README summary

Insomnia is a comprehensive API development tool with core capabilities including API design (with a native OpenAPI editor), debugging, testing (via native test suites), API mocking, and building CI/CD pipelines with its CLI. It offers flexible storage options including local, Git sync, and cloud storage, making it ideal for developers who need to debug, test, and collaborate on APIs.

Why it matters

The project is a well-known tool in the API development space, feature-rich and open-source, supporting a wide range of protocols and flexible storage options, suitable for full-stack developers and DevOps engineers.

TypeScript

#15

Lightricks/LTX-2

Official Python inference and LoRA fine-tuning package for the first DiT-based synchronized audio-video generative foundation model.

7,678

Stars

1,220

Forks

+196

Today

README summary

This repository is the official Python package for the LTX-2 model, enabling inference and fine-tuning. It describes LTX-2 as the first DiT-based audio-video foundation model capable of synchronized generation, high fidelity, and multiple output modes. The README provides detailed instructions for model downloads, environment setup, and various inference pipelines, making it suitable for developers and researchers interested in cutting-edge synchronized audio-video generation.

Why it matters

The project showcases a frontier direction in synchronized audio-video generation with a novel architecture and comprehensive capabilities. The inclusion of a full toolchain, including LoRA training, makes it highly valuable for technologists focused on multimodal generation and content creation tools.

Python