GitHub 2026-10-26 Weekly Trending Projects

A coding-agent skill for security engineers that uses six phases and independent verification instead of one-shot scanning.

Ruflo provides enterprise multi-agent orchestration with self-learning RuVector, for automating dev workflows, audits and security governance.

Role-focused customizable plugins to integrate tools and codify workflows for Claude, improving team consistency.

OpenCode is an open-source AI coding agent for terminal and desktop, model-agnostic and LSP-enabled, designed for local or remote-driven development workflows.

Orca offers local multi-agent AI orchestration for engineering teams, using worktree isolation and parallel agent comparisons to speed development and review; however, evaluate maintenance and compliance risks due to unclear repo activity, missing license information, and reliance on external agent subscriptions.

Hyperframes renders deterministic videos from plain HTML, combining AI-agent-driven authoring and adapter-based animations to power automated content pipelines and repeatable video production.

Adds SKILL.md workflows to Android agents, focusing on development tasks where LLMs underperform.

Graphify converts code and documentation into a traversable, provenance-aware knowledge graph using local tree-sitter AST parsing for interactive AI-assistant queries and audits; non-code semantic enrichment requires external models, and the license/community status should be verified before adoption.

Provides long-term session memory and handoff across tools and agent vendors, enabling seamless context continuity for developer teams and automated workflows; license, maintenance, and privacy controls require review.

ComfyUI is a visual, modular node-based engine for visual creators and production pipelines; it supports local offline execution and multimodal models to build reusable workflows for images, video, audio and 3D.

Anthropic-maintained Claude Code plugin directory offering standardized plugin structure and easy installation; useful for developers and operators, but third-party plugins require careful security and license review.

vLLM is a high-throughput, memory-efficient inference and serving engine for large language models that supports multi-GPU deployment, FP8 quantization, and zero-overhead prefix caching; it targets online and batch inference scenarios with emphasis on runtime performance and scalability.

Provides teams using Claude Code or general LLM coding agents with a compact, verifiable set of four guidelines and integration options to reduce assumptions, prevent overengineering, and limit edit scope to lower regression risk.

T3 Code is a minimal web GUI and cross-platform desktop for Codex/Claude coding agents, enabling quick validation of agent workflows; it is very early-stage with limited maintenance and contribution paths, suitable for exploratory trials and proof-of-concept.