A curated library of machine-readable task prompts and Agent Skills for coding agents, focusing on common agent failures like broken setup scripts, vague issues, and tests requiring unavailable services. Includes fixtures with planted defects for validation.
Open source. Open possibilities.
Discover quality open-source projects, submit projects anonymously, and claim and edit your own project.
A little curiosity. A world of open source.
THE FIRST COLLECTIONLess than 10 stars, 5+ commits in the last 30 days — projects made with care.
Lumen is a private and lightweight web browser with its own engine, written from scratch in Rust without using Chromium or WebKit wrappers.
Trahens is a research protocol for privacy-oriented rendezvous route discovery and control-plane management in decentralized, path-aware networks.
SunAgent is a modular, provider-independent AI agent runtime built from scratch to explore core components like agent loops, tool calling, and memory. It features a declarative harness system, run tracing, and evaluation benchmarks for understanding and extending modern agent architectures.
MOSAIC is a synchronized multi-camera and audio recording suite designed for research labs to capture social and activity interactions.
go-whatchanged prints API diffs between two versions of a Go module, with read-only operation, release-aware version selection, CI-friendly outputs, and a GitHub Action for pull request comments.
Assura is a Rust command-line tool that validates repository structure, naming, file, section and Markdown conventions against a single local policy, giving fast advisory review, repairable feedback for AI coding agents, and a check gate before merge.
SessionReplaySDK is a privacy-focused iOS SDK for recording user sessions. It captures video, touch events, logs, and network requests with automatic PII masking, supporting SwiftUI/UIKit and local-first storage.
A standard library of interfaces for local tool capabilities like downloads, storage, and job management, featuring a conformance suite to test implementations.
Gemaal is a Kubernetes test cluster housekeeping service that automatically cleans up ephemeral installations. It acts as a 'pump' to drain unused resources, ensuring shared test environments remain stable and habitable by managing TTLs, ring-pair teardowns, and orphaned artifacts without blocking client operations.
Threadbase Mobile is a local-first mobile control plane for monitoring and managing AI coding agent sessions, such as Claude Code and Codex, via a connected tb-streamer instance.
debux is an open-source container debugger for Docker and Kubernetes that debugs distroless, scratch, and minimal images with a ready-to-use toolbox. It attaches to targets, shares namespaces, and exposes the filesystem for inspection.
A Docker-based XHProf traces viewer for PHP application profiling, built on Alpine Linux with multi-architecture support.
Touchstone provides guidance prompts and CLI tooling for solo developers managing AI agents across GitHub projects. It enforces constrained workflows via GitHub policy, ensures state legibility, and carries consistent rules. The CLI validates checks, verifies claims, and sequences PR operations (open, status, merge, answer) with documented recovery paths.
ChaosSQL is a deterministic concurrency and isolation fuzzer for PostgreSQL, MySQL and SQLite. It schedules concurrent transactions, detects invariant violations such as lost updates and write skew, shrinks failures to minimal reproductions, and can guard CI pipelines against regressions.
ripr is a static mutation-exposure analysis tool that identifies changed behavior your tests reach but don't actually verify, without running mutation testing.
Build-Eye is a CI monitoring and root cause analysis system for vLLM-Ascend. It automatically monitors GitHub builds, classifies failures (code, infrastructure, interference), generates fix recommendations, and archives standardized reports to streamline debugging.
Pixel FX Forge is a browser-based procedural generator for creating animated pixel-art ability effects, allowing users to design and export game-ready VFX.
Kokopu-React is a React-based library designed to create and display chessboards and related chess components.
Ambition is an experimental, ECS-native 2D platformer engine built on Rust and Bevy. It features a headless-testable architecture, agent-native authoring tools for sprites and music, and a data-driven design focused on long-term elegance and reusability.
An open-source AI coding assistant platform with workspace-aware analysis, GitHub integration, team collaboration, a capability-based plugin system, and read-only Stellar/Soroban developer tooling including contract inspection, XDR decoding, and project detection.
A Lean 4 formalization of the convex-analytic and variational-analysis results from the book "Variational Analysis" by Rockafellar and Wets.
Phronesis is a RETE rules engine that enforces project conventions for LLM agents outside the context window, preventing contextual drift in long sessions.
A repository tracking HDF5 test results across major High-Performance Computing (HPC) systems, including Frontier and Perlmutter, via CDash.
TraceFrugal is a local Go application that visualizes token usage for Claude Code and Codex, offering MCP result packing, cost regression gates, and optimization trials to help developers manage AI context efficiency and verify savings.
PilotDeck subtask acceptance and local repair plugin enhances AI agent delivery quality via structural checks, independent read-only model review, and bounded in-session repairs. It supports delivery contract validation, white-box experience memory, and fault injection demos to ensure atomic task failures are auto-repaired without affecting passed sibling tasks.
AgentLeak is an open-source privacy testing tool for AI agents. It detects data leaks across tool calls, memory, and logs using a local Python SDK, CLI, and MCP tools, providing redacted reports and CI gates without cloud dependencies.
Zentra is an AI-powered CLI security scanner for developers, combining SAST, DAST, supply-chain, API, and IaC analysis with LLM orchestration, CI integration, and an authorized pentest mode.
QA Studio is an open-source, self-hostable test management platform for QA teams. It offers hierarchical test suites and cases, test run planning across environments, rich dashboards for results and trends, media attachments, and a full REST API.
Whitaker provides opinionated Dylint lints for Rust to enforce readable code, sensible module sizes, and avoiding panics in production. Includes nine standard lints (e.g., banning expect() outside tests, module line limits) and configurable thresholds. Supports multiple languages for diagnostics.
Netsuke is a build-system compiler that transforms YAML and Jinja manifests into static Ninja build graphs for fast, incremental execution.
Liferay AI Commerce Accelerator (AICA) is a suite of client extensions that uses generative AI to rapidly generate and deploy commerce data—catalogs, accounts, orders—into Liferay DXP. It offers a zero-dependency CLI, real-time monitoring, dependency-aware generation, and provider-agnostic AI support (OpenAI, Gemini, Claude).
PSIE Genesis Kernel v3.2.2 is a multi-agent governance framework implementing 476 encoded ethical laws based on mathematical geometry rather than traditional morality. It ensures non-aggression, consent, and diversity through modular architecture and a rigorous audit process requiring 90% consensus for integration.
A collection of helper functions designed for AWS Lambda functions developed using TypeScript.
RecurSpec is a Node.js command-line testing tool that checks whether the recovery instructions in error messages actually resolve the original failure, running each case in an isolated temporary workspace and blocking unsafe commands.
A collection of TypeScript helper functions designed for integrating nRF Cloud APIs within AWS Lambda functions.
Divebell is an extensible web development and debugging toolkit for coding agents. It connects page context, browser capabilities, and existing development tools, enabling agents to reproduce, diagnose, and verify web application issues directly in real scenarios via a CLI and extensions.
Agent Interlock is an agentic AI security framework that declares the actors and trust boundaries of an AI agent system, wraps each actor's tools through an SDK or proxy, and enforces MCP tool threats M1–M9 across observe, shadow and enforce stages, recording every verdict in an append-only ledger.
Antiserum is a local, open-source CLI scanner for text training data that flags possible planted poison such as trigger n-grams, label flips, duplicate injects, paraphrase overweight, hidden Unicode, instruction overrides, and known signature matches. It runs offline, writes deterministic receipts, supports CI/SARIF, and lets users confirm findings and contribute public signatures.
Browser‑based real‑device lab for Android, iOS and Tizen. Provides a web UI to manage and automate mobile device testing from a single workspace, with Docker and native host support.
A browser-based tool that converts images into high-detail text art using Unicode braille glyphs, featuring advanced dithering and edge detection.
fingerjoin aggregates WebFinger responses from multiple fediverse applications on one Kubernetes domain. It forwards queries to annotated HTTPRoutes, merges JRD results by priority, and returns standard status codes per RFC 7033.
PyFlow is a research-oriented static analysis framework for Python, offering IRs, program analysis, optimization passes, security checking, supply-chain analysis, and CLI/LSP/MCP tooling.
Prettier plugin for Mojolicious .html.ep and .html.epl templates. Formats HTML with Prettier and embedded Perl with perltidy. Supports standard Prettier options and .perltidyrc. Idempotent and degrades without perltidy.
Sophia Stack is a modern, atomic X11 desktop system designed to replace ambient trust with explicit authority boundaries and isolated namespaces.
SkillGuard is a local-first security scanner for coding-agent skills and MCP servers. It detects prompt injection, credential theft, data exfiltration and unsafe downloads, then blocks unscanned or risky tools via Claude Code and Codex hooks.
A TypeScript library that builds 3D terrain in Babylon.js using Geospatial Information Authority of Japan tile elevation data. It provides terrain viewer and diorama APIs as an npm package, including WebXR support and numerous demos.
A generator-free reactive SQLite layer for Dart and Flutter that provides typed tables and live streams without requiring build_runner or code generation.
Zeekr Shortcut (Car Version) is an experimental Android app for the ZEEKR 7X head unit that turns the four surround-view cameras into a 2×2 dash cam, an electronic rear-view mirror window, and a QR-code file transfer tool.
DashLord is an automated dashboard for technical best practices for websites and APIs. It uses GitHub Actions to scan URLs and generate a static web report published via GitHub Pages.
A Pascal-focused project family providing a UI framework, schema generation tooling, and developer utilities.
AI Infra Bench evaluates frontier AI models on real-world AI infrastructure engineering tasks, starting with vLLM and covering CPU and GPU workloads including bugs, features, performance changes, refactors, and tests.
AutoFFF is an open-source Python tool that automatically generates FFF (Fake Functions Framework) fake function definitions for C API header files, simplifying unit testing of embedded C code by eliminating manual work writing and maintaining test headers for platform API faking.
Mybox is a workstation configuration and tool management system designed to handle software installations without requiring superuser access.
SMTP Tool is a portable Windows desktop application for testing, diagnosing, and debugging SMTP servers. It supports TLS 1.3, real-time open/click tracking via local HTTP server, latency benchmarking, and interactive protocol terminal.
SUM is an open-source tool for verifying AI text transformations with cryptographic receipts. It checks what changed, what was preserved, and what was lost when AI rewrites text, with cross-runtime Ed25519 signatures and offline verification.
A multi-platform storage enclosure and disk visualization tool providing physical slot views and health monitoring for TrueNAS, ESXi, Linux, and other hardware.
Zero-dependency LÖVE2D framework with scenes, input, saves, audio, UI, ECS, physics, and in-game scene editor. Includes reference game Gem Haul. Targets LÖVE 11.5, no third-party libraries.
Lintro is an MIT-licensed command-line tool that brings many linters, formatters, type checkers and security scanners together under one interface, with consistent reports, auto-fixing, Docker images and optional AI-assisted review using your own API key.
An open-source Minecraft server load tester that connects swarms of bots and reports real server-impact metrics such as estimated TPS, latency, throughput and kick reasons. Includes CLI, web GUI, control-plane API, debug console and programmable bots. Authorized testing only.
GitHub profile README for Sachin, a self-described applied AI architect and engineer. It points to selected open-source projects covering agents, retrieval (RAG), evaluation and reliability tooling, and lists skills, working style, engagement terms and contact details.
Socket Page Automation is a dependency-free state machine for browser page automation. It manages anti-bot challenge cadences in npm human-verification flows by pausing operations via an injected seam instead of blind-retrying, optimizing batch operations within cooldown windows.
An Unreal Engine plugin for in-game bug reporting that captures gameplay video and logs for submission to the BetaHub platform.
Socket Security Skills provides standardized AI agent skills for dependency security tasks, including vulnerability scanning, package review, patching, and secure updates, compatible with 40+ coding agents.
Python library for generating podcast RSS feeds. Supports Apple Podcasts, Overcast, Pocket Casts, and PSP Certified players. Create feeds and episodes with metadata.
A Python CLI and MCP server that reads a git diff against a CodeGraph index and returns a short ranked list of the user journeys a change could have broken, instead of re-running every test. Recall-first, driven by a journey registry and guarded against stale indexes.
Prebuilt CDK Terrain (cdktn) bindings for the HashiCorp Google provider, enabling infrastructure as code for Google Cloud using various programming languages.
Development repository for Autoplot containing scripts demonstrating bugs (in 'bugs' directory) and feature requests (in 'rfe' directory). Includes dated folders for quick bug demos. Scripts can be run directly in Autoplot via URL.
LiveDoc is a living documentation platform that brings Gherkin-style BDD syntax to Vitest and xUnit, with a real-time web viewer for test results, tag filtering, data tables and scenario outlines.
RESOLVE is an LLVM-based security tool designed for automated vulnerability triage and remediation. It uses enhanced software bills-of-materials (eSBOMs) and program analysis to identify and address CVEs.
Avogadro 2 plugin and CLI tool for computing and visualizing Intrinsic Bond Orbitals (IBOs) via Psi4. Supports occupied and valence-virtual IBOs with Pipek-Mezey localization, degeneracy resolution, and detailed analysis. Closed-shell systems only.
Conductress is a Valkey benchmarking framework that queues and runs performance, memory, and replication tests. It offers a TUI, CLI, and statistical analysis for comparing results. Supports local and remote server testing with configurable parameters and repetitions.
A tool for mapping Discord web client chunks to readable code paths and identifiers to aid reverse engineering.
A GitHub Actions-based automation pipeline that daily collects recruitment notices from 42 major Korean hospitals using Selenium, verifies data integrity via a RaiT gateway, and generates filtered Excel reports.
Web UI Quality is an Apache-2.0 open-source preview CLI that keeps scope, actual change, evidence and verification separate for AI-assisted Web edits, and refuses to treat a model suggestion as write authority.
A bit-exact Minecraft Java Edition overworld generator implemented in pure C11, offering significantly higher performance than the vanilla server.
pgfathom is a read‑only CLI tool that discovers undocumented PostgreSQL relationships by analyzing catalog data, usage evidence, and statistics. It outputs verdicts (broken, confirmed, weak) and reviewable SQL scripts to help fix hidden foreign keys.
Laravel package integrating Ranetrace for error tracking, privacy-first website analytics, uptime and performance monitoring, centralized logging, and MCP server access for AI agents.
Tarnished’s Arsenal is a Windows desktop optimizer for Elden Ring builds that calculates optimal weapon, affinity, and stat combinations.
A Terraform provider for the Frostmoln Cloud Platform, allowing users to manage cloud resources as code within EU/EEA datacenters.
NovaFabric is an open-source, self-hosted CLI that captures any AI or HPC run — script, agent, or model call — as a portable, secret-redacted, replayable evidence capsule you own, with no application code changes.
Iris is a local-first MCP server that scores AI agent runs for quality, safety, and cost using 20 built-in deterministic rules, an optional LLM judge, and a web dashboard — all on your machine with no account or telemetry.
aiHelpDesk is a governed AI agent system for production database incident response and remediation across VM, Docker/Podman, and Kubernetes deployments. It diagnoses failures, executes fix playbooks under audit trails, and learns from every resolved incident.
Reprobuild unifies build, dev environment, and infrastructure into one compile-time-verified graph. It auto-discovers dependencies, enables hot reloading across projects, and ensures reproducibility via hermetic builds and state-based nodes.
SecureFlow is a local-first application security platform in Rust. It combines deterministic source-to-sink analysis, human validation workflows, and offline API inventorying. The tool prioritizes reproducible evidence, maintaining strict separation between automated candidates and final human judgments, with no automatic vulnerability validation.
A pnpm-workspace monorepo providing flat ESLint 9+ configurations and related tooling published under the @williamthorsen scope.
PARTHA turns one repository revision into a sealed, queryable intelligence model — building architecture graphs, dependency graphs, engineering reviews, insights, documentation, and optional AI context from a single persistent snapshot with evidence and provenance.
An intelligent ORM analysis and optimization platform for Django that detects N+1 queries, duplicate queries, and slow queries.
A DeepSeek Harness plugin that exports OpenTelemetry traces, metrics, and logs to GreptimeDB, with seven Grafana dashboards for tracking token usage, costs, and performance of AI agent runs.
CodexUserData 是一款 Windows 桌面悬浮窗工具,用于可视化本地 Codex 的 Token 用量、模型分布及账号额度。支持多种悬浮形态、自定义 HTML 界面及历史趋势分析,帮助用户实时监控 AI 编码助手的使用情况。
Pulse is a batteries-included Spring Boot observability starter providing cardinality firewall, timeout-budget propagation, SLO-as-code, async context, PII masking, and stable error fingerprints with zero agents and one dependency.
LogSquirl is a fast, open-source log explorer for Windows, macOS, and Linux. Search multi-gigabyte files, follow live logs, and chart patterns. Built with Qt6/C++, it's a fork of klogg and glogg.
cctrace is a TLS-intercepting tracer for Claude Code, Codex, Grok, and Kimi CLIs that captures every API call in a live web UI with session replay, cost tracking, and cross-project dashboards.
A reusable, pluggable Python/pytest-based test suite for validating ADBC database drivers. It runs SQL queries against drivers to test feature support, data type handling, and database-specific quirks, generating documentation of supported and unsupported features. Supports per-driver test overrides and a consolidated .txtcase test format.
Peaky Peek is a local-first, open-source audit and trust console for AI agents. It captures causal chains, verifies claims deterministically, and produces explainable trust scores with failure narratives — all running on your machine.
ping6.it is a web UI for comparing IPv4 vs IPv6 latency and reachability using Globalping and RIPE Atlas probes. It runs ping, traceroute, DNS, and HTTP measurements on the same probes for fair side-by-side comparison.
Web of Flaws is a structured, markdown-first catalog of web security vulnerabilities and secure code replacements designed for developers, security engineers, and AI coding agents.
An adapter that allows ASGI applications to run on an RSGI server.
dsnitch is a single-binary, zero-configuration terminal UI tool powered by Linux eBPF that provides real-time network and DNS egress inspection for Docker containers, attributing outbound connections directly to container names without modifying container networks.
ctop is a top-like interactive terminal interface for real-time container metrics. It supports Docker and runC, offers multi-host monitoring, secure web dashboards, and low-overhead performance with no external dependencies.
recoverage is a command-line tool designed to track code coverage changes over time and prevent coverage regressions within CI pipelines.
Warden Desk is a read-only, eight-agent rule engine for on-chain risk analysis around Solana launchpad coins: seven rule agents screen candidates and an eighth Warden agent can only refuse, logging verdicts with contract addresses and reasons while never executing trades.
CoalLedger is a docs-health checking suite for AI coding agents — the documentation counterpart to CoalMine's code canaries. Seven canaries flag wrong, incomplete, stale, contradictory, broken, unreadable and (optionally) leaked documentation, with a bundled zero-dependency CommonMark+GFM AST engine, context-judged severity and choice-gated fixes.
A Java starter library designed to facilitate the execution of integration tests using TestContainers and WireMock.
A Java-based starter library designed to facilitate the storage of files in AWS S3.
A DuckDB compatibility harness for rudb. It verifies conformance through differential execution against a real DuckDB binary, weighted SQL coverage, storage format round trips, and C ABI checks.
The Construct is a disposable Ubuntu VM for unattended AI coding agents like Claude Code and Codex. It runs inside a Hyper-V sandbox on Windows, ensuring host isolation while providing full root access to agents. Features include project profiles, remote hosting support, and integrated T3 Code client.
A privacy-focused, client-side web utility for parsing server access logs to analyze bot traffic, crawl budgets, and infrastructure egress costs.
Krkn Operator is a Kubernetes-native platform for centralized multi-cluster chaos engineering, enabling teams to orchestrate and manage chaos experiments across Kubernetes and OpenShift clusters from a single control plane.
qyl is a local OpenTelemetry investigation stack for .NET. It allows developers to instrument applications with minimal code, run a collector locally, and visualize traces, logs, and metrics via an embedded dashboard or API.
RetroChess is a Chess plugin for RetroShare 0.6 that lets you play chess with RetroShare friends and contacts. It installs as a shared-library extension and adds a chess game window plus chat invitations, and ships with a standalone Qt board debugger for testing chess rules.
A Node.js monorepo providing a collection of universally useful utility functions.
VibeCoder is an automated GitHub issue worker that monitors repositories, picks up issues, writes code using AI agents (Claude, Codex, Gemini), runs quality checks, and opens pull requests. It supports review feedback loops, self-healing, and cost optimization.
A 3D software rasterizer built from scratch in C99, supporting multiple platforms including desktop and WebAssembly.
A Steadybit extension that discovers AWS resources and injects faults into various AWS services for chaos engineering and reliability testing.
BGSTM is a methodology-agnostic software testing framework built on six core phases, from test planning to results reporting. It ships guidance, templates, worked examples, and a self-hostable FastAPI + React reference application with traceability and quality dashboards.
MCP Arcade is a command-line testing tool that runs a small catalog of experiments, called atoms, against MCP servers, judging results by the JSON-RPC wire and sandbox file changes rather than by tool descriptions or model prose.
AgenTerm is a Rust-native terminal and local AI fleet controller for Windows, macOS, and Linux, featuring hierarchical tabs and native automation capabilities.
A Pulumi Python-based infrastructure-as-code project for managing a Talos Linux Kubernetes cluster spanning OCI and a homelab LAN with Cilium networking.
A TypeScript-first programmatic wrapper for the tmux CLI, supporting Node.js and Bun runtimes.
wicked-garden is a curated skill catalog for coding agents (Claude Code, Codex, etc.) that fills gaps in verification, memory, and multi-model review. It re-derives 'done' from evidence, tracks injected edges, and enforces rigorous gates without forcing a specific workflow.
A personal Elixir/OTP game-automation bot for a 2D MMORPG, using screen capture and synthetic input on macOS. It features a supervised process fleet, pure decision logic, and fail-safe actuation controls.
TUnit-native math assertion library for .NET: fluent tolerance, sequence, statistics, linear algebra, number theory, and 3D geometry assertions generated into TUnit's Assert.That pipeline; AOT-compatible and reflection-free.
Ditto scans JavaScript/TypeScript repositories with ts-morph, fingerprints each function through an LLM, clusters behavioural duplicates by embedding, then executes pure functions in a sandbox to show where they diverge.