Recently Evaluated

A research hub for the paper and survey "Toward Efficient Agents: A Survey of Memory, Tool learning, and Planning," with organized paper navigation across memory, tool learning, and planning topics.

Hapsland is a developer tool that gives coding agents realtime semantic review on code changes, with privacy-aware request scoping and feedback loops back to the agent.

Porch is an open-source macOS work log for coding agents. It tracks Claude Code and Codex sessions, summarizes work by project and time period, and can save Markdown notes for tools like Obsidian.

A research-oriented optimizer project claiming to cut AdamW optimizer-state memory for LLM training by 50% through factorized variance and frequency-domain compression. The repo includes a Python POC, benchmark claims, and a stated drop-in usage pattern for PyTorch training.

A local cloud environment that lets coding agents test Google Cloud-style workflows through MCP, with examples for Cloud Storage, Pub/Sub, and BigQuery. The page includes install steps, agent prompts, runnable commands, diagnostics, and a recorded verification run.

A macOS CLI for Firebase Crashlytics that helps AI coding agents and developers search production crashes, inspect stack traces, and measure crash impact from the terminal.

Rinkata is a shared project context layer for teams using coding agents, with knowledge capture, goal tracking, decision handling, and evidence packs for completed work.

A Hugging Face paper page describing a memory layer for LLMs that stores KV state to encrypted NVMe and reloads it later without recomputing, with results from 50M-token experiments on Gemma models.

Aura is a self-hosted, multi-user AI agent in Go with temporal graph memory, scheduled jobs, MCP tools, and multi-channel access through CLI, web, Telegram, and WhatsApp. The repo shows a full product surface with Docker Compose deployment, local/provider model options, memory, retrieval, and sandboxed operator workflo

An iPhone and iPad SSH terminal built for monitoring and responding to AI agents, with sealed agent notifications, Secure Enclave keys, and live panes for workspaces and tabs.

An open-source Pac-Man benchmark where AI decision models compete in real time against scripted ghosts, with a public leaderboard, replayable games, and instructions for benchmarking your own endpoint.

A macOS CLI and skill for Claude Code/Codex that lets AI agents draw arrows, boxes, and text on the screen to guide a human during prompts, approvals, and demos.

An agent-facing service marketplace for discovering capabilities, requesting work, paying for quote-based services, and verifying results with receipts. The page highlights MCP access, multi-agent interoperability, and several service families including verification, intelligence, risk, and transactions.

Search2o is a Python package for running searchable AI agents inside an organization, with a REST API, bundled GUI, and support for self-hosted or vendor LLMs.

An MCP server for background computer use by agents. It can start app sessions, drive windows with mouse and keyboard actions, capture screenshots, inspect UI trees, and run scripted recipes across macOS, Linux, and Windows.

Open-source agent skills for Claude Code, Cursor, Codex and other coding agents that audit SaaS stacks, compare vendors, and surface plan-limit, billing, overlap, and compliance issues using live vendor pricing pages.

Botscent is a lightweight library that detects AI agents visiting a site and can identify which agent is behind a request when evidence is available. It includes both a server-side signal check and a small client-side script, with docs, FAQ, and GitHub links shown on the page.

Favz indexes public GitHub repos to show which MCP servers, skills, plugins, and agent configs people actually set up. It tracks rising tools, compares stars with real usage, and includes a CLI/plugin angle for monitoring your own setup.

Singularity is a developer tool that gives coding agents procedural memory so repeated tasks can be handled with fewer tokens and less setup. It includes a CLI, installer scripts, hooks for multiple agents, and workflow records that learn from past sessions.

A self-hosted, local-first chat workspace for OpenAI-compatible models with built-in MCP tools, web search, and long-term memory.

Official implementation of LittleBit and LittleBit-2 for sub-1-bit LLM compression via latent factorization and quantization-aware training.

Local-first Node tool that reads Claude Code and Codex logs to price AI coding-agent usage by repo, model, or client. It also supports billing markups, JSON output, and an MCP server for asking agents what their work cost.

A Mac app that centralizes engineering alerts from tools like GitHub, Sentry, Jira, Linear, Datadog, and Vercel, then lets you hand issues off to AI coding agents such as Claude Code, Codex, or Cortex Code.

VoiceGremlin is a CI/CD testing tool for AI voice agents that places real phone calls, evaluates results through an API, and returns pass/fail outcomes with transcripts.