GitHub Trending Weekly · Week 43, 2026
Security-audit skills and agent orchestration took over last week as developers moved coding agents from demos into audited, parallel work.
Agent security · security-audit-skill runs multi-phase audits with verified findings, while rea links coding agents to binary and .NET analysis tools to trace features.
Agent orchestration · orca runs many agents in parallel git worktrees with phone monitoring, and ai-memory keeps shared long-term context across harnesses.
Codebase context · graphify builds a queryable local knowledge graph so agents navigate code, and opencode offers a terminal agent with full model control.
Weekly star gains
All languages · Daily leaderboard totals
Biggest star gains
Projects ranked by the star gains recorded for this issue.
- 1morluto/rea+78,562
Connects a coding agent to local binary, JS/Electron and .NET analysis tools so you can trace how a feature works.
- 2
A coding-agent skill that runs a multi-phase security audit with independently verified findings for codebases you own.
- 3stablyai/orca+9,828
An orchestration desktop for running many coding agents in parallel git worktrees, with phone monitoring and remote execution.
- 4Graphify-Labs/graphify+6,868
Builds a local, queryable knowledge graph of a codebase so coding agents can navigate it instead of grepping.
- 5anomalyco/opencode+6,857
An MIT-licensed terminal coding agent for developers who want full control over model choice and local file access.
- 6heygen-com/hyperframes+5,633
Turns HTML, CSS and seekable animations into deterministic MP4s for developers and agents producing video programmatically.
- 7Comfy-Org/ComfyUI+4,648
Node-graph engine for running open diffusion image, video and audio models locally with fine control over every step.
- 8akitaonrails/ai-memory+3,541
Shared long-term memory for coding agents so you can switch harnesses or machines mid-task without re-explaining context.
- 9ruvnet/ruflo+3,241
A harness that wraps Claude Code and Codex with swarms, persistent memory, and cross-machine federation for multi-agent work.
- 10
A curated plugin marketplace for Claude Code, giving developers a vetted place to find and install extensions instead of hunting GitHub.
New projects
New projects ranked by their recorded total stars.
- 1Gathin23/cleanyourtv34,436
Strips ads and preinstalled apps from Android TV and Google TV over Wi-Fi, then installs a clean home screen launcher.
- 2deepseek-ai/DeepInfer22,324
Serves DeepSeek's own inference stack to teams running DeepSeek-V4.1-Flash at production scale on NVIDIA or Ascend hardware.
- 3Zagitta/gpu-unlock6,057
Linux kernel module that unlocks full FP16/BF16/TF32 tensor-core throughput on GeForce RTX 20/30/40 cards before the NVIDIA driver loads.
- 4
Runs the Switch Zelda games natively on Apple silicon Macs, iPhones and iPads using your own game files.
- 5pengchujin/aerial-tv4,956
A clean, ad-free replacement home screen for China-market Android TVs, with tvOS-style layout and automatic takeover at boot.
- 6
Runs your own Switch copy of Diablo II: Resurrected natively on phones, Macs and HarmonyOS devices with rebuilt touch controls.
- 7
Replaces the ad-filled smart TV home screen with a clean, custom app launcher for Google TV, LG and Samsung sets.
- 8HeyPuter/everything3,867
An MIT-licensed monorepo rebuilding everyday proprietary apps — spreadsheets, docs, chat, notes — as web tools for people who want self-hostable alternatives.
Daily leaders
The top eligible project by star gains on each recorded day.
- debpalash/VoiceStudio+2,082
A local, open-source voice studio for cloning, designing, dubbing, transcribing and narrating audio without a cloud account.
- stablyai/orca+1,559
An orchestration desktop for running many coding agents in parallel git worktrees, with phone monitoring and remote execution.
- DietrichGebert/ponytail+2,051
A skill pack that pushes coding agents to reuse existing code and write only what the task needs.
A coding-agent skill that runs a multi-phase security audit with independently verified findings for codebases you own.
A coding-agent skill that runs a multi-phase security audit with independently verified findings for codebases you own.
A coding-agent skill that runs a multi-phase security audit with independently verified findings for codebases you own.
A coding-agent skill that runs a multi-phase security audit with independently verified findings for codebases you own.