gawkbot

gawkbot is the open source Grok Bot.

claude-code-model-proxy

Use the Codex model inside Claude Code!

muhyeop-full

A Claude skill that writes Korean in a pure martial arts fiction style.

deslop

A professional plain English style guide and agent skill that translates legalese and purges AI slop.

XYZZY

A self-hosted collaboration space that records technical decisions by people and AI with a chain of evidence

It lets multiple people and AI agents develop and review ideas in parallel in one space, then publish decision documents linked to evidence and records that can reveal changes.

Ziplex

A tool that compresses local projects into structured context AI can read

Developers passing large codebases to AI coding tools can analyze and compress file structures and dependencies, then create a small aif.json context reviewed by a person before it is saved.

claude-code-arcade

A tool that turns the Claude Code status bar into an autoplaying arcade

Developers running Claude Code for extended periods can see work status as animated games like Ms. Pac-Man, R-Type and Frogger, and change themes or control the games when they want.

tubeviz

AI-directed, beat-synchronized music video generation from YouTube footage, with semantic analysis, interactive editing, and GPU-accelerated rendering.

saccade

An AI agent interface that verifies semantic changes and action results in an approved browser tab

For tasks where automation becomes unreliable after the first click, such as signed-in admin pages or dynamic forms, it helps AI agents keep track of the state of a user-approved Chrome or Edge tab.

sonde

A local context engine that answers code structure questions within a token budget

It indexes repositories as symbol graphs and returns results through 3 MCP tools, helping AI coding agents understand call relationships, change impact, and related tests without repeated searches.

kavranta

A management app that keeps real environment variables local while reducing secret exposure during AI work

A local-first desktop tool for finding, editing, linking, sharing, and deploying real environment variable files across projects, while helping users avoid pasting protected values into AI chats.

hedgemony

A checker that detects nonexistent Python names and code contract violations without another AI

It uses the interpreter and package registry to check AI-written Python code for unpublished packages, nonexistent methods, disallowed arguments, and contradictions in executable examples.

DROS-VajraClaw-Hacker

A Docker security gateway that controls AI agent access to the operating system with deterministic policies

It helps independent developers and AI researchers put a security boundary between autonomous agents and the local operating system and control execution with Markdown or YAML policies and W3C DID-based permissions.

preseason

An open-source benchmark measuring which development tools LLMs recommend for real web app questions

Helps developers and researchers choosing development tools compare recommendations from multiple models on fixed web app prompts and review tool and service rankings and methodology.

Xazz

A DSL for building AI pipelines in a single Rust script

Xazz is a Rust based DSL for developers who want to combine Polars preprocessing, Burn deep learning compilation and static security guardrails in one script for an AI pipeline.

Hillock

A lightweight neurosymbolic memory engine that runs offline on edge hardware

A neurosymbolic memory engine for local AI developers who want long-term memory and document similarity checks on edge devices without a large vector database or a generative LLM for document processing.

Astera

A desktop workbench for coordinating long-running parallel work in Claude Code and Codex

Helps developers running multiple AI coding sessions schedule and resume work, assign dependent tasks to Claude Code and Codex, and run them safely in parallel in separate Git worktrees.

oh-my-subagents

A local multi-agent runtime for operating a persistent, accountable AI team

oh-my-subagents is a local runtime that persistently manages parent-to-subagent delegation in Codex and Claude, overseeing assignments, waiting, and result collection so work context survives interruptions.

memnest

An MCP memory service that lets multiple AI coding tools search the same local memory

It helps developers using multiple MCP clients, including pi, Claude Code, and Codex, store decisions and conversation history on their own devices across sessions and workspaces, then search them when needed.

tokensift

A local static-analysis linter that finds wasted tokens in LLM prompts and payloads.

Developers seeking to reduce LLM API costs and context usage can analyze prompt strings, message arrays, and tool schemas at the tokenizer level to find waste.

Itsuki

An open-source memory layer shared across AI tools

A shared memory layer that helps people working across AI tools extract and reuse structured memories from 26 connected tools with a single connected key.