Explore the directory

AgentLens is a trust-first marketplace and workspace for AI agents where sellers publish an agent for others to rent under a time-limited contract, buyers compare execution boundaries and audit evidence, and a six-dimensional audit framework verifies trust claims.

• Agent frameworks
• AI evals

J-Space Cognition Suite is a model-agnostic, inference-time control system distributed as an installable Skill that organizes an agent's working context into a managed workspace for deep reasoning, tool use, verification and recovery, without changing model weights.

• Agent skills
• Agent memory

A Claude Code and Codex agent skill that turns articles, chart screenshots and reports into labeled illustrations, redrawn charts and reference-based diagrams in a consistent style.

• Agent skills
• Generative media

AIHelms is an enterprise platform for managing an organization's AI resources: a unified gateway across providers such as OpenAI, Azure, Claude and DeepSeek, per-department budgets and token-level cost attribution, and registration of internal MCP tools and Skills.

• LLM gateways

Arcade MCP is an open-source Python framework for building MCP servers and tools, with a decorator letting a tool declare the OAuth scopes it needs so Arcade handles authorization and token refresh without exposing the token to the client or model.

• MCP servers

Tel-Agent is a self-hosted, open-source gateway connecting an existing phone line to an AI agent over SIP, checking each caller against routing rules before passing the call to a person, blocking it, or handing it to an agent that can transfer calls and take messages.

• Speech

GrapeRoot is an open-source launcher that builds a semantic graph of a codebase's files, symbols and call chains, then pre-loads the relevant code into an AI coding assistant's prompt, instead of letting the assistant discover context turn by turn.

• Agent memory

Vigolium is a vulnerability scanner that combines a fast, deterministic native scan across 317 modules with an agentic scan mode where an AI plans attacks, selects modules, writes custom extensions and triages results against a codebase.

• Security

Agentica is a TypeScript framework for AI function calling that turns a list of functions, whether a TypeScript class, a Swagger/OpenAPI document or an MCP server, into an AI agent, automatically compiling the function-calling schema instead of requiring it to be written by hand.

• Agent frameworks

Nemo Skills is NVIDIA's collection of pipelines for improving large language models: synthetic data generation, model training, and evaluation across benchmarks in math, code, science and instruction following, scaling from a workstation to a Slurm cluster.

• Models
• AI evals

deja is a Go binary that gives coding agents shared memory by indexing session history that Claude Code, Codex, Cursor and 30-plus other agents already write to disk, handing the right one back to whichever agent asks, with no model call or capture step.

• Agent memory

The Microsoft 365 Agents SDK lets developers build agents once and deploy them across channels such as Microsoft 365 Copilot, Microsoft Teams, and custom web apps, using whichever AI services or models they choose, with separate client libraries for C#, JavaScript and Python.

• Agent frameworks

A from-scratch reimplementation of the vLLM inference engine with self-contained paged and flash attention, built to benchmark attention mechanisms during prefill and decode.

• Inference
• Tutorials

Optio is a self-hosted platform that runs AI coding agents as durable, observable sessions on Kubernetes pods or a team's own laptops, then wires them into scheduled reports, webhook responders, ticket-to-merged-PR pipelines and long-lived agents that message each other.

• Agent frameworks

Copilot API is a local AI gateway that exposes GitHub Copilot, a built-in Codex provider and configurable third-party model providers through one OpenAI- and Anthropic-compatible endpoint, so tools such as Claude Code, OpenCode and Codex can all talk to the same local server.

• LLM gateways

jevos is an open-source, self-hosted alternative to a hosted yes/no decision service called Jev: it answers a yes/no question about a piece of text on a laptop CPU in 50 to 220 milliseconds and speaks the same wire format so existing client code keeps working.

• Inference

Harden.Run AIF is a local safety layer for coding agents that installs native hooks in tools such as Claude Code, Codex, Cursor and Gemini CLI to block risky actions before they run, with a local dashboard and anonymous, code-free operational telemetry.

• Agent skills
• AI evals

Optimum Quanto is a PyTorch quantization backend for Hugging Face's optimum library, now in maintenance mode, that inserts quantization and dequantization automatically and supports int2, int4, int8 and float8 weights and activations across CPU, CUDA and MPS devices.

• Models
• Libraries

A set of Kotlin programming challenges with tests and multiple worked solutions, intended for practicing coding skills and preparing for technical interviews.

• Tutorials

MemSlides is a hierarchical, memory-driven agent framework for personalized presentation slides that separates persistent preferences from working memory and tool experience, editing only the smallest affected slide region during a revision instead of the whole deck.

• Agent frameworks
• Agent memory

rizzo-pii is a small, CPU-friendly model that detects 22 categories of personal data in Italian legal text, including Italian tax and property identifiers, and drives a local, reversible anonymize-then-restore workflow so sensitive documents can still be sent to a cloud LLM.

• Models
• Security

An embeddable vector database for Go with a Chroma-like interface and no third-party dependencies, running in memory with optional persistence.

• Databases

NeoHorse is a family of open-weight language models sized for agent workflows, with 4B and 9B versions tuned for tool use, coding and instruction following, plus a companion decision model that uses prefill-only inference to pick an action or score a condition.

• Models

Knowledge Agent Template is an open-source, forkable starting point for an AI agent that answers questions from a knowledge base using grep, find and cat inside a sandbox, with no embeddings or vector database, deployable as a chat, GitHub or Discord bot.

• Agent frameworks
• RAG