2,419 tools and products — trending open source, and what gets used in AI and other work.
Scene Card Studio is an AI visual director and visual-narrative engine for turning personal photographs into structured, editable narratives rather than applying a simple style transfer. It separates observable photo evidence from narrative interpretation and direction, then processes photos through Scene Cards, a Narrative System, a Prompt Compiler, image generation, aesthetic review, and retry or acceptance. The output includes directed images, deterministic layouts, versioned generation prompts, and hash-linked review records. Its Narrative Systems define story structure separately from replaceable Expression Profiles that control surface language.
pi-peer is a pi extension for discovering and messaging coding-agent sessions. It provides `list_peers` to show sessions, working directories, and statuses, and `message_peer` to send short plain-text messages to a named session. Messages carry no conversation history or files and arrive marked as peer text with no authority: they cannot approve actions, change configuration, or execute slash commands. Local messages use disk-backed mailboxes that survive restarts and queue mail for inactive sessions; sessions can also communicate across team machines through a shared Redis server. The extension supports inbound-message policies and stores local peer data with restricted filesystem permissions.
actions/checkout is a GitHub Actions action that checks a repository out into the workflow’s `$GITHUB_WORKSPACE` directory for use by subsequent steps. It can select a repository and ref, perform shallow or complete-history fetches, sparse or partial checkouts, and fetch submodules, Git LFS files, tags, and authenticated content. By default, it fetches only the commit for the event’s triggering ref or SHA; `fetch-depth: 0` retrieves all branches and tags. Authentication can use a token or SSH key configured in the local Git configuration and removed during post-job cleanup; recent releases store persisted credentials in a separate file under `$RUNNER_TEMP`. If Git 2.18 or newer is unavailable, the action falls back to the GitHub REST API to download files. Checkout v7 migrates the action to ESM, updates direct and transitive dependencies, and refuses fork pull-request code by default for workflows triggered by `pull_request_target` or `workflow_run`, which have access to the base repository’s token, secrets, and runner. Reviewed exceptions can opt in with `allow-unsafe-pr-checkout: true`.
KTransformers is an open-source research framework for efficient large-language-model inference and fine-tuning through heterogeneous CPU-GPU computing. Its user-facing capabilities are provided through the kt-kernel source tree: high-performance inference and supervised fine-tuning (SFT). For mixture-of-experts models, it uses heterogeneous expert scheduling, placing frequently used or “hot” experts on GPUs and other experts on CPUs with optimized AMX, AVX512, or AVX2 kernels and NUMA-aware memory management. It supports CPU-side INT4/INT8 quantized weights, GPU-side GPTQ, native BF16 and FP8 precision, and multiple CPU, GPU, and NPU backends. The SFT capability integrates with LlamaFactory for hybrid CPU-GPU fine-tuning of large MoE models, including LoRA and full-parameter training, with INT4/INT8 quantization options.
OpenCut is a free, open-source video editor for the web, desktop, and mobile, positioned as a local alternative to CapCut. It provides basic video-editing capabilities while keeping users' video files on their devices. The project is being rewritten around a Rust core intended to support browser, desktop, and mobile clients from one codebase. The planned architecture includes an Editor API, third-party plugins, an MCP server for AI agents, headless batch rendering, and an in-editor scripting tab. The existing classic version remains available through opencut.app while the rewrite is developed at new.opencut.app. OpenCut is licensed under the MIT License.
Hiring Agent is an open-source Python resume-to-score pipeline. It extracts structured data from resume PDFs, enriches candidate information with GitHub signals, and evaluates projects, work experience, and technical skills with supporting evidence to produce an explainable evaluation. The repository describes it as a ranking aid for prioritizing resume review, not an applicant-tracking system or a product for HackerRank customers. It includes local-model configuration, with Gemma 4 listed as the default demonstration model, and provides command-line setup and usage instructions.
oh-my-pi is an open-source terminal coding agent and fork of Pi by Mario Zechner, with an IDE-oriented interface wired into the agent. It supports context-aware code edits, code search, language-server operations such as renames, debugging through debugger operations, persistent Python and Bun code-execution workers, web browsing, sub-agents, and MCP; the project describes support for dozens of model providers and includes built-in tools. It runs on macOS, Linux, and Windows, with installation options including a shell installer, Homebrew, Bun, Nix, and PowerShell.
Klaat Code is an open-source, terminal-native AI coding agent from KlaatAI. It reads and edits project files, runs shell commands, performs type checks after edits, and asks for permission before risky operations. It supports planning, sub-agents, MCP, hooks, and skills, and indexes projects into a code knowledge graph containing symbols, callers, callees, and semantic-search data for targeted exploration. Its hosted Klaatu-o1 router classifies each request and dispatches it across six model-cost tiers—nano, fast, code, reason, heavy, and titan—rather than using one model for every task. The CLI is a thin client to this service; the repository states that routing decisions, model health tracking, pricing, and the code-graph index run server-side. It supports providers and models including Claude, GPT, Gemini, and DeepSeek, and can be installed with npm.
Usertour is an open-source user onboarding platform for creating in-app product tours, checklists, launchers, surveys, and related onboarding experiences. It supports user targeting, custom styling, multiple environments, version tracking, analytics, and self-hosting. The platform includes a built-in MCP server that lets AI assistants such as Claude Code, Cursor, or Codex create onboarding flows and surveys from natural-language instructions. The documented self-hosting path uses Docker Compose, and the project provides a web application, documentation, and an SDK for integrating onboarding experiences into products.
Alexandrie is a self-hosted, open-source knowledge base and wiki for teams and individuals, positioned as an alternative to Notion, Confluence, and Obsidian. It uses an extended CodeMirror 6 Markdown editor with support for LaTeX, custom containers, interactive checkboxes, snippets, speech-to-text, and embedded diagram creation. Notes are organized through multi-tenant teams, workspaces, categories, and nested documents, with full-text search, a command center, integrated Kanban boards, public links, and node- or workspace-level permissions ranging from no access to owner. It is distributed as an offline-first progressive web app with synchronization, OIDC/SSO support, Docker deployment, backup and restore pipelines, and S3-compatible media and backup storage.
Frontman is an open-source, browser-based AI coding agent for visual frontend editing. It lets users select an element in a running application, describe a change in plain language, and edit the underlying source files with hot reload. The agent uses the live DOM, component tree, computed CSS, routes, source maps, and server logs to identify the relevant code, then presents the resulting diff for review. It supports frontend projects using frameworks and tools including Next.js, Astro, Vite, React, Svelte, and Vue, and is distributed under Apache 2.0 / AGPL-3.0 as listed in its repository.
ComPDF Self-Hosted is an open-source PDF editing, conversion, and document-processing platform from the KDAN ecosystem. It is deployed privately with Docker Compose and provides a browser-accessible tool center for merging, splitting, rotating, watermarking, encrypting, decrypting, and otherwise editing PDFs, as well as converting between PDF, Office formats, HTML, TXT, CSV, RTF, JSON, Markdown, OFD, and image formats. Its image-conversion functions include image-to-document and image-to-PDF workflows, with OCR-related capabilities indicated by the repository topics. A management console handles API keys, API usage, license status, operation-log auditing, account management, and system configuration; commercial support and an enterprise upgrade are also offered.
tududi is an open-source task and project organizer for life and work, available as a self-hosted application or hosted subscription. It organizes tasks, subtasks, projects, areas, notes, and tags in a hierarchy, with filtering by status and due-date categories, project sharing, responsive access, and an installable progressive web app with offline caching and queued synchronization. Its recurring-task system supports daily, weekly, monthly, weekday-specific, and last-day-of-month patterns, custom intervals, completion-based recurrence, linked parent and generated tasks, and optional end dates. The project also provides Telegram capture, CalDAV synchronization, OIDC, and a REST API.
Dashwise is an open-source, self-hosted homelab dashboard that combines links, search, RSS news, notifications, wallpapers, glanceables, widgets, and integrations with self-hosted services in configurable pages. It includes built-in authentication and SSO/OIDC support, a Spotlight-like search opened with Ctrl+K, and an HTTP endpoint for receiving push notifications by topic. Dashboards are configured through the user interface rather than configuration files; page configurations and user preferences are stored in PocketBase and read by the React frontend through its page-configuration API. Integrations are defined as YAML files that provide widgets, glanceables, and shortcuts. The application uses a React single-page frontend bundled with Bun, a Bun/Hono API layer, and a PocketBase backend, and can be deployed with Docker Compose. The repository warns that the project was partly rewritten after its last major release and recommends backing up the PocketBase directory before upgrading.
tuicr is a terminal user interface for code review with Vim key bindings. It presents a GitHub-style continuous diff, allowing reviewers to scroll through changed files in one stream and add comments at line, range, file, or review level. Review status is tracked at file or hunk granularity and persisted across sessions. It reviews uncommitted changes, commit ranges, and pull or merge requests from GitHub, GitLab, and Bitbucket, using Git, Jujutsu (jj), or Mercurial repositories. Reviews can be submitted to GitHub, GitLab, or Bitbucket, copied as structured Markdown, or written to standard output. The project is implemented in Rust and can be installed through its install script, Homebrew, Pacman, Cargo, Mise, Nix, or pre-built binaries.
ESP32 Bit Pirate is open-source ESP32-S3 firmware that turns supported development boards into multi-protocol hardware development and analysis tools, inspired by the Bus Pirate. It provides a command-line interface over USB serial or Wi-Fi web serial for sniffing, sending, testing, and scripting wired protocols such as I2C, SPI, UART, 1-Wire, CAN, GPIO, JTAG, and USB, as well as wireless technologies including Bluetooth, Wi-Fi, Sub-GHz, RFID, infrared, LoRa, and FM. The firmware includes protocol-specific tools such as EEPROM and flash access, UART bridges, device scanning, signal recording and replay, radio analysis, PWM and direct I/O control, and Bus Pirate-style bytecode or Python scripting. The project supports several ESP32-S3-based boards, including ESP32-S3 Dev Kits, M5Stack devices, and LILYGO boards, and provides web tools, hardware guides, recipes, and documentation through its project website.
Project AIRI is a self-hosted AI companion for creating virtual characters or digital pets, with Live2D or VRM avatars and real-time voice chat. It connects to multiple model providers and can interact through Discord, Telegram, and games including Minecraft and Factorio. The project supports web, macOS, and Windows, and provides installation options through Winget, Scoop, and Homebrew Cask.
Santa is an open-source binary and file access authorization system for macOS, developed by North Pole Security. Its system extension monitors executions and file access, evaluates requests against a local database, and records events; a GUI agent notifies users about blocked operations, a background service synchronizes configuration with a remote server, and a command-line utility manages the system. In MONITOR mode, binaries are allowed unless blocked while launches are logged; in LOCKDOWN mode, only listed binaries may run. Rules can use code-signing attributes such as CDHash, certificate, TeamID, or SigningID, as well as regular-expression-based paths, with rules evaluated by specificity. Allowed binaries are cached, and userland components authenticate one another over XPC by checking matching signing certificates.
Octane is an open-source JavaScript UI framework and successor to Inferno, developed by Dominic Gannaway. It uses a compiler to turn React-style components into direct DOM code before they ship, removing the virtual DOM and manually maintained dependency arrays while retaining familiar APIs such as useState, useEffect, memo, context, portals, Suspense, and transitions. Components can use standard JSX or Octane's .tsrx syntax, which adds template directives such as @if, @for, @switch, and @try. The compiler derives hook dependencies from captured values, tracks hooks by call site rather than call order, and rejects hooks in plain JavaScript loops; keyed @for is provided for per-item hook state. The runtime uses delegated native DOM events, plain ref props, compiled render paths, and an LIS-based keyed reconciler. The native renderer does not include class components, Server Components, or a synthetic event system.
BlazeDiff is an open-source diff ecosystem for JavaScript applications and other runtimes. It provides pixel-perfect image comparison, Rust and SIMD-based native and WebAssembly diff engines, structured image-diff interpretation, object diffing, perceptual metrics including SSIM, MS-SSIM, Hitchhiker's SSIM, and GMSD, and a from-scratch Rust PNG codec. The ecosystem includes JavaScript, Rust, Python, and command-line packages, along with framework-agnostic and React UI components for visualizing differences. Its agentic visual-regression tool discovers routes, captures deterministic screenshots, and sends ambiguous diffs to a coding agent for a verdict.
Webiny is an open-source, self-hosted content platform and headless CMS that runs serverlessly inside the user's AWS account. Its TypeScript framework provides a GraphQL API, custom content models, field-level permissions, localization, versioning, lifecycle hooks, dependency injection, admin UI and infrastructure extension points, and supports AI-assisted development through an MCP server. The platform includes a visual website builder with a Next.js SDK, a file manager for digital assets, publishing workflows with drafts, reviewers, scheduled publishing and audit trails, and native multi-tenancy with isolated data, users, assets and permissions. It runs on AWS Lambda, DynamoDB, S3 and CloudFront, with infrastructure provisioned through Pulumi; optional OpenSearch and VPC configurations are also supported. Webiny is extended through code and can be deployed using open-source infrastructure templates rather than managed servers.
Exo is an AI agent harness for recursive self-improvement, providing tools, tasks, integrations, and a layered architecture that separates executive policy, protected state, and sandbox execution. It can inspect its own code and runtime logs, incrementally modify prompts, memory, tools, and harness policy, clone itself, and manage a lineage of clones. An event log remains outside these modifications as a canonical record intended to prevent recursive loops.
CoachAI is an iOS fitness app presented at coachai.tech that uses an iPhone’s sensors and AI-powered, on-device movement analysis to provide guided workouts, exercise coaching, and feedback.
mpai is a small GitHub-hosted project (presented on a GitHub Pages site) that positions itself as a collaborative AI coding tool, describing the concept as "AI coding is multiplayer now" and using the tagline "Open your teammate." The project appears to be created and published by the GitHub user godfaddaai.
MascotAI is a web application that generates animated SVG mascots for web and mobile applications. It offers customizable gestures and themes, and produces download-ready mascot asset packs for direct integration into products.
Airtop is a platform that automates Google Ads campaign management. Its Google Ads offering builds, monitors, and optimizes campaigns from a conversation, providing keyword research, campaign creation, waste audits, and performance reporting.
A payment "skill" from UniwebPay that enables AI applications to generate payment links and begin accepting payments via prompt-driven commands. It is intended to reduce the onboarding, KYC, integration, and infrastructure work needed for monetizing AI agents by returning payment links/tokens and related payment responses. Documentation and the skill endpoint are presented at the UniwebPay skill site.
OpenAnalytics is an open-source, self-hostable web analytics stack developed by OpenLabs. It uses a lightweight browser tracker without cookies, fingerprinting, or cross-site profiles, and provides aggregate-only analytics for page views, custom events, sessions, web vitals, funnels, retention, embeddable widgets, public sharing, live presence, and Stripe revenue attribution. The repository is a pnpm monorepo comprising a tracker, ingestion collector, queue worker, API, ClickHouse query gateway, realtime SSE service, Next.js dashboard, and CLI. Postgres stores control-plane data, ClickHouse stores events and rollups, and Valkey provides the event queue and realtime cache. The query gateway is the only component permitted to read ClickHouse and verifies Ed25519-signed query envelopes; visitor identity uses a salted hash that rotates nightly, while optional city-level geolocation is resolved against a database on the host. It also includes CSV/JSON import and export, an MCP server, and an AI assistant. The project is licensed under AGPL-3.0. It can be self-hosted with Docker Compose, or used through the authors' hosted instance at getopen.so. Release images are published to GitHub Container Registry.
Human Review is a local visual review tool and agent skill for editing HTML and Markdown files, reviewing localhost pages, and sending feedback to an AI coding agent. It opens a file or URL in a browser, where users can edit text and basic formatting, rearrange or remove page elements, resize or paste images, and attach comments to exact phrases, images, charts, or sections. Direct HTML edits and image changes save automatically; Markdown and localhost reviews are sent to the agent, which applies the changes to the source and refreshes the page. The tool includes CLI commands for review sessions, polling, status, and setup, plus a local server, editing and feedback SDK, browser client, Markdown renderer, and instructions for agent harnesses such as Claude Code and Codex. It runs on the user's computer without an account, cloud service, database, or API key.
ASD-STE100 Skill is a Claude Code skill that rewrites dense or ambiguous English into ASD-STE100 Simplified Technical English for AI-agent outputs, including tool descriptions, error messages, READMEs, and inter-agent instructions. It offers Strict mode for procedures, error messages, and tool descriptions, and STE-flavored mode for explanatory prose. The skill reads the input for meaning, flags sentence-level issues such as ambiguous wording, complex tense, unclear passive voice, multiple instructions, long sentences, noun clusters, phrasal verbs, nominalizations, semicolons, hedge stacks, and marketing adjectives, then rewrites flagged sentences while retaining facts, conditions, and scope qualifiers. It can also provide a before-and-after table naming the rules that each sentence breaks.
pdfcn is a free, open-source library of customizable PDF components for React, designed in the style of shadcn/ui. It uses a registry format and CLI workflow for copying components into an application, where they can be customized and composed into tables, forms, charts, invoices, reports, and other documents. PDFs can be rendered through either Takumi or Forme. The project provides zero-configuration defaults, live previews in its documentation, shared themes and props, and composable React components with renderer primitives. It is licensed under the MIT License.
tokentab is a local tool that reads Claude Code session logs and converts token counts into cost reports organized by model, project, day, and activity. It provides JSON output and a local dashboard, with the reported data remaining on the user's machine.
An open-source, executable playbook of agent skills for U.S. utility-patent work. It provides SKILL.md instructions, checklists, intake questionnaires, examiner and adversary protocols, and deterministic Python checks for dates, claim counts, and fees. Its three main workflows are a pre-filing audit, a simulated USPTO prosecution examination loop, and an adversarial design-around test. The repository grounds its simulations in 35 U.S.C., 37 CFR, the MPEP, and current USPTO materials that the agent is instructed to refresh, but it does not file applications, contact the USPTO, provide legal advice, or perform acts reserved for registered practitioners. It is licensed under GPL-3.0.
ha.mr is a static URL compressor and QR code optimizer that runs entirely in the browser without a back-end database. It detects common URL components such as protocols, www prefixes, and index.html, encodes ports as numeric values, matches domains and top-level domains against a Huffman-coded dictionary, and compresses the remaining URL segments using predefined character sets or Huffman coding. For QR codes, it uses the QR alphanumeric character set to reduce encoding overhead.
Cinematic Zoom is a browser-based three.js application that flies a camera from orbit to selected locations on Earth and records the movement as an MP4. Users choose a landmark or enter a name, coordinates, DMS coordinates, or a map link, then select a shot preset—Dive, Descent, Orbit, Flyby, or Hyperzoom—a lighting setup, and a duration. The camera's distance, bearing, pitch, lens, and roll each follow separate curves rather than a single interpolation, while the app positions the sun for the selected location and renders Google's photorealistic 3D tiles. The project is built on the three.js 3d-tiles-renderer, with tiles streamed through Cesium Ion and atmosphere and clouds supplied by @takram/three-geospatial. It requires Node 22.12 or newer for local development and a browser with WebGL 2; recording additionally requires WebCodecs with a hardware video encoder. A Cesium Ion token or a Google Maps Platform key with the Map Tiles API enabled is required, and the app stores the token in the browser.
ttfx is a dependency-free Rust command-line tool that applies terminal text animations to piped input and distributes them as a single static binary. It is a parity-exact port of the Python TerminalTextEffects project, preserving its effects, animation engine, command-line interface, cursor behavior, options, and implementation quirks rather than introducing new artwork. Users select effects such as decrypt, beams, matrix, or a random effect with commands such as `cat banner.txt | ttfx beams`; the project documents 37 effects and byte-identical output to the Python original for the same input and seed. The port and original are MIT licensed.
BlueFerry is experimental Linux software that connects a paired iPhone to a Linux desktop over Bluetooth, without a Mac relay, Apple login, cloud service, or subscription. It provides native GTK, KDE/Kirigami, Quickshell, and terminal clients for reading and sending SMS, RCS, and iMessage, starting conversations, syncing contacts, marking messages read, and optionally mirroring other iPhone notifications. Local message history can be encrypted with GNOME Keyring or KDE Wallet. It only receives messages observed while connected and does not download an iCloud Messages archive. Attachments, reactions, typing indicators, FaceTime, calls, and complete sent-message history are unsupported; group replies are disabled or require confirmation when Bluetooth does not provide enough participant information. The project distributes native packages for several Linux distributions and warns that changes to iPhone Bluetooth behavior could affect compatibility.
A framework-free WebGL2 package for rendering Apple-inspired translucent glass surfaces for folders, rectangles, pills, and circles. Its renderer models screen-space refraction using a squircle signed-distance-field thickness field, Snell refraction with a meniscus rim, variable-blur scattering, Fresnel reflection, chromatic dispersion, edge highlights, contact shadows, hit testing, and live backdrops. It provides the original frosted V1 renderer and a separate clear optical V2 renderer, each with its own material contract; both can be used together. The package is distributed through npm, requires WebGL2, accepts CSS-pixel element coordinates, and can load image backdrops.
MiniMax Music 3 is a music-generation model from MiniMax AI that creates complete songs of up to five minutes from lyrics and a music description. Lyrics can use section tags such as verse, chorus, bridge, instrumental, solo, and outro; the music description specifies style, emotional progression, vocals, instrumentation, arrangement, and production. A structured-caption format can organize these instructions into global metadata, vocal details, and arrangement sections. The model uses a hierarchical autoregressive hybrid architecture: an 8B Global LLM predicts the first residual-vector-quantization codebook frame by frame to model long-range musical structure, while a 0.6B Local LLM predicts the remaining acoustic codebooks within each frame. Its continuous hidden-state synthesis module combines the two models' final hidden states with Flow Matching and Flow-VAE to produce 32 kHz, 16-bit stereo WAV audio, preserving information for vocal articulation, instrumental texture, and temporal continuity.
Agent-Safe Pipeline is a Decionis TypeScript library and runnable reference implementation for routing AI-agent actions through an independent authorization boundary. It captures an agent’s proposed action, target, and parameters as an immutable intent, sends the intent to a Decionis policy gate for an ALLOW, ESCALATE, or BLOCK decision, and coordinates verified human approval followed by policy re-evaluation for escalated actions. Agents cannot authorize their own actions, access downstream privileged credentials, or select trusted handlers. SafeExecutor accepts only the captured intent and decision, uses a sealed ActionRegistry to map action names to trusted handlers and validate parameters, and consumes a single-use grant bound to the intent before calling an API. The repository includes examples for Shopify refunds, GitHub deployments, procurement, and governed MCP tools, plus canonical-hash conformance vectors and architecture, threat-model, and security-evidence documentation. It is a self-hosted reference implementation rather than a hosted authorization service; its documented safety boundary also depends on provider-side identity, least privilege, network isolation, and incident response. The demos require Node.js 22.14 or later and pnpm 9.
oss-pr-reviewer is an AI-powered command-line tool for reviewing GitHub pull requests and producing structured Markdown or JSON reports for maintainers. It fetches pull-request metadata and changed-file patches with Octokit, normalizes reviewable content, skips binary, patchless, truncated, and oversized files with explanations, and splits large text diffs into deterministic batches without cloning or executing the repository code. The tool sends batched changes to OpenAI for structured findings, validates responses with Zod, then merges, deduplicates, orders, and filters findings by severity. It can apply repository-specific review rules and ignored paths from a trusted base-branch `.oss-pr-reviewer.yml` file, retry transient GitHub and OpenAI failures, and optionally publish bounded advisory reports through GitHub Actions. It requires Node.js 20 LTS or newer and, according to the repository, is run from a checkout because its package-ready release is not yet published to npm.
MCP-Memory is a Model Context Protocol server that gives AI coding agents persistent memory across chat turns and sessions. It stores memory records as human-readable Markdown documents following the Open Knowledge Format (OKF v0.2), with YAML frontmatter, hierarchical index files, and an update log, while maintaining a local SQLite FTS5 index for key lookups, keyword search, tag filtering, and namespace-scoped retrieval. The server exposes MCP tools for storing, retrieving, searching, and deleting memories, and supports project, user, and default namespaces; a setup wizard configures supported MCP clients including Claude Desktop, Cursor, Antigravity, Windsurf, and Codex.
pi-transcribe is a Pi extension for local speech-to-text dictation. It records microphone audio through a configurable Pi terminal shortcut, processes it with a locally downloaded speech model, and inserts the transcription at the editor cursor; streaming-capable models process audio in roughly 500 ms chunks, while other models process the complete recording after capture stops. It also registers a transcribe_file tool for local audio and video files and a /transcribe command for configuring languages, models, microphones, and shortcuts. File transcription requires FFmpeg, shares a loaded model across queued jobs, admits at most two file operations at once, runs one FFmpeg decoder at a time, and limits decoded audio to 128 MiB.
An experimental community preset for DeepSeek Harness, an LLM-agent harness. It starts a session with the real Minimal tool schema containing only a shell tool and a file-reading tool, using that condition to anchor the model's initial trajectory. After the first durable tool call or assistant reply, it promotes the session to the full Standard tool catalog, making heavier tools available on demand. The repository also includes related zero-anchored, self-introduction, prefab, and eternal-Minimal modes, and states that it is not an official DeepSeek preset or affiliated with DeepSeek.
Multi-Agent Workbench is a local-first browser control room for orchestrating coding agents such as Claude Code, Codex CLI, and generic PTY-based command-line agents. Agents work in task rooms under editable YAML role cards that define permissions, speaking style, and decision boundaries; a rule-driven orchestrator pauses execution and risky commands for human approval, while persistent sessions expose terminal activity and status. The system uses a local event-driven architecture with a React interface, Fastify backend, REST API, server-sent events, agent adapters, an approval service, and a SQLite event store. Every action passes through the API and is recorded in an append-only event log, whose events are projected into timelines, decisions, terminal views, artifacts, and diff views. It also provides task templates, Markdown reports and decision logs, real-time room updates, and configurable risk resolutions. Code-writing agents are intended to use isolated Git worktrees, with diff and patch review before changes reach the main workspace; the repository marks this worktree and merge flow as in progress.
Colite is an experimental C++17 structure-from-motion and multi-view stereo system that reconstructs 3D scenes from overlapping images. Its pipeline performs incremental structure-from-motion with support for multiple models, continuation, and Sim(3) merging; feature extraction and matching; PatchMatch stereo with geometric consistency; stereo fusion; meshing; simplification; and texture-atlas generation. It produces COLMAP-compatible sparse models, RGB and normal dense point clouds, Poisson or Delaunay meshes, and textured OBJ models with MTL and atlas PNG files. CPU and optional CUDA implementations are provided for SIFT, descriptor matching, and PatchMatch Stereo, and resumable stage manifests allow processing to continue from saved stages. The repository describes the current build as alpha-quality, primarily developed and tested on Ubuntu 22.04 and WSL2; native Windows builds are not yet validated as a supported release target.
Ember is a collection of redshift-aware color palettes for terminals, interfaces, charts, and heat maps. It models warm-screen filters at 3400 K, 2000 K, and 1200 K, where each RGB channel is multiplied by a different gain, and supplies palettes designed to preserve text readability, categorical color separation, contrast, perceptual separation, and ordered lightness after the modeled transform. It includes neutral or warm-neutral text and surface colors, terminal accents, categorical palettes, and 256-sample sequential maps; deeper filters intentionally provide fewer distinct semantic colors and may assign unsupported ANSI names as aliases. The palettes are distributed in terminal, CSS, JSON, Python, and Matplotlib formats. The repository documents deterministic filtered-output simulations based on published gains rather than photographs, calibrated physical color temperatures, or predictions of every display pipeline.
TermDOM is a JavaScript/TypeScript library for building terminal applications with HTML, CSS, and DOM APIs. It renders DOM nodes into a grid of terminal character cells, supporting CSS stylesheets and the box model, flexbox and table layout, text-width handling for CJK and other characters, forms, scrolling, Web Components, selection, and keyboard, mouse, focus, paste, and input events. It implements a browser-like rendering pipeline against character cells rather than pixels: styles and layout are recomputed for mutated content, the result is painted into a cell buffer, and differences from the previous frame are written to standard output as ANSI escape sequences. Input escape sequences are decoded into events dispatched to DOM nodes, and utilities such as querySelector(), MutationObserver, ResizeObserver, and getBoundingClientRect() operate on the cell-based layout.
Graft is an open-source codebase-context tool for coding agents, developed by NanoNets. It builds a repository graph and stores the result as linked Markdown files, with one node for each system, API, or concept and links describing how parts of the codebase connect. The graph supplies targeted context to agents instead of requiring them to rediscover repository structure during every task. It provides a CLI and MCP server, with integrations for Claude Code, Cursor, Codex, Gemini, and other coding agents. The CLI supports repository initialization, graph building, targeted search and orientation through commands such as `graft grep` and `graft map`, and graph visualization. `graft init` can install the agent wiring and background rebuild hooks; the generated graph is a local, regenerable cache that is added to `.gitignore` by default.
lumabri is a pure-C system for running large mixture-of-experts models across a swarm of peer machines using the colibri engine. A machine hosting a model serves it to other machines; model bytes needed during inference are fetched from peers on first use and retained in a shared content-addressed store. With Segment available, peers retain assigned layer ranges and their state while clients receive the tokenizer, embeddings, final transform or head, and conversation state; the system can instead fall back to the expert/CAS engine or local execution when a complete Segment route is unavailable. It supports CPU- and SSD-first participation as well as GPUs, and provides serving and chat commands, a terminal UI, configurable disk and compute donation, machine and health diagnostics, swarm monitoring, and low-priority peer-to-peer or signed relay transfers. The video also describes routing of missing model blocks or expert activations between peers, with hashes, signatures, and optional encryption used for protection.
mcptoon is a cross-agent Model Context Protocol (MCP) management CLI that gives agents access to configured MCP servers through ordinary command-line calls, rather than loading large tool catalogs into their context. It discovers existing server configurations, maintains ~/.mcptoon/config.json as a single source of truth, synchronizes that configuration across agents, serves a tool-name manifest on demand, and composes a single-entry proxy for MCP access. Tools are launched when invoked, and the project supports native MCP JSON, structured output, optional TOON compression, and envelope passthrough; the videos also describe output checks for prompt injection and credential leaks. The project is distributed as a Python package for Python 3.10+ and is described as using only the standard library, with no third-party dependencies. Its repository documents compatibility with agents including Claude Code, Cursor, and Codex, and provides commands such as `quickstart`, `demo`, and `call`. It is free and open source under the Apache-2.0 license.
Grok2API is a self-hosted, Go-based API gateway with a built-in React administration console. It manages separate account pools for Grok Build, Grok Web, and Grok Console, synchronizes account credentials, quotas, and models, and provides unified OpenAI- and Anthropic-compatible APIs with account and model routing. The gateway exposes Grok capabilities including chat, image generation and editing, and video. The project is intended for technical research and learning and asks users to comply with Grok's terms of use and applicable laws.
celld is an open-source daemon from Deno that runs Cloudflare Workers and Durable Objects on self-hosted machines. Each Durable Object is a named cell with its own SQLite database; cells store long-term state in an S3-compatible, Google Cloud Storage, or Azure Blob Storage bucket. A celld node embeds V8 and executes Wrangler bundles. Nodes sharing a bucket form a fleet, using conditional bucket writes for cell ownership and signed peer HTTP for routing and replicated-log transport, without a membership protocol, failure detector, or consensus service. Committed SQLite writes are captured as LTX data; nodes persist that data locally and upload it to the bucket, while takeovers recover open logs from the previous owner before restoring a cell. celld can run locally with `celld dev` and a local object store, or as a container for Linux x86-64 and ARM64. Its installer provides a verifiable binary release, and deployed Worker projects require esbuild on the PATH.
RustDesk is an open-source remote desktop application written in Rust and designed for self-hosting as an alternative to TeamViewer. It provides remote screen control, file transfer, clipboard synchronization, and TCP tunneling across major desktop and mobile platforms. Users can use the project's rendezvous and relay server, operate their own, or write a custom rendezvous/relay server; the project provides downloadable binaries and build instructions.
ToolJet is an open-source low-code platform for building and deploying internal tools, dashboards, business applications, workflows, and AI agents. Its visual builder uses drag-and-drop components and supports multi-page apps, collaborative editing, JavaScript and Python logic, a built-in no-code database, and connections to databases, APIs, cloud storage, SaaS applications, and other data sources. The community edition can be self-hosted with Docker, Kubernetes, or major cloud providers, while ToolJet AI adds natural-language app generation, AI-assisted query building and debugging, and an agent builder. The project also provides access control, encrypted data handling, plugin and connector extensibility through the ToolJet CLI, and hosted deployment through ToolJet Cloud.
OpenSEO is an open-source SEO tool developed as a self-hostable alternative to Semrush and Ahrefs. It provides workflows for keyword research, rank tracking, competitor insights, backlink analysis, site audits, and AI visibility, using a DataForSEO API key for the underlying SEO data. The project exposes an MCP server and reusable agent skills that let agents such as Claude Code, OpenClaw, and Hermes access SEO data and perform guided SEO tasks. It can be self-hosted with Docker for personal use or Cloudflare for internet-facing use across devices or teams; a hosted version is also available, with pricing described as pay-as-you-go for API usage and a subscription option.
An open-source agent skill and standard-library Python service for stripping AI provenance marks from text and files owned by the user. The skill is a thin HTTP client that invokes a separately running service, so the agent host does not need Python. Deterministic processing handles invisible Unicode characters, exotic spaces, bidirectional and tag characters, while an agent rewrite pass and optional rewrite_text.py hook address statistical token-sampling watermarks. File processing targets C2PA, EXIF, XMP, and document properties across image, document, ebook, web, video, and audio formats, including PNG, JPEG, WebP, AVIF, HEIC, BMP, GIF, TIFF, SVG, PDF, DOCX, XLSX, PPTX, EPUB, ODT, HTML, Markdown, MP4, MOV, M4A, M4V, WAV, MP3, and FLAC. The repository describes support for class-level provenance ecosystems including Claude, Gemini/SynthID-Text, OpenAI provenance surfaces, Kirchenbauer-style green-list marks, and keyed-Gumbel/EXP marks. It includes installable skills for Claude Code, Cowork, Cursor, and other supported hosts; the installer uses Python 3.10+ with no external dependencies.
React Email Editor is an open-source React component developed by Unlayer for embedding a drag-and-drop email editor in an application. It lets applications load and save email designs as JSON, listen for editor events, customize the editor configuration, add custom blocks, support translations, and export responsive HTML. The component is distributed through npm and renders client-side in the DOM; its README documents support for React 16.8 or later and Node.js 18 or later in version 2. It also offers an optional AI assistant.
Soloop is a startup offering a “one-person company OS” for solo founders, independent makers, and indie founders. It provides an AI-based founding team and AI-powered tools and workflows intended to help users grow and monetize the products they build.
Nitro API (Nitro Translate) is a human translation API for AI agents, provided from nitrotranslate.com. It connects AI agents to professional native‑speaker translators in 80+ languages and advertises an account‑free, no‑API‑key agentic payment flow via MPP.