Botscent

Detect signs and types of AI agents visiting a website.

A lightweight library that helps site operators examine requests and page behavior for signs of AI agents, then use the detected type and reason to guide their own response.

kora-doctor

A CLI that finds potentially unnecessary LLM calls in AI agent execution records.

It helps developers reviewing AI agent execution costs analyze AUDR records and find possible waste to examine in tool use, repeated reasoning, context growth, and other areas.

neptune

A free, open-source terminal built in Rust, with workspaces, split panes, and tools for working with AI coding agents.

Recly

Skip the $159 recorder: one shortcut on the watch you already wear, free on-device transcription, notes by the AI you already use.

cuesheet

Pick MSW mock scenarios, response delay, the clock and offline mode from a link or a panel, so AI agents, tests and QA all see the same screen.

SCM

A local search app for finding photos and video scenes on a Mac using visual information and text.

For Mac users whose photos and videos are scattered across folders, it shows search results using image similarity, OCR text, and spoken dialogue from videos.

life-forge

Autonomous flight simulator for AI agents — co-evolutionary adversarial red-teaming with 3D MAP-Elites quality-diversity.

claude-routine-cleanup

Bulk-delete old Claude Code scheduled-task (routine) runs safely: keeps the newest N and every run you replied in.

valet

Self-hosted cloud coding agents: Claude Code and Codex in per-thread sandboxes, one docker compose - dropalltables/valet

alps-writer-plugins

Plugins for coding agents to write and manage product requirements and design decisions.

For teams documenting product development with Codex or Claude Code, it provides two plugins covering requirements writing, architecture decision records, implementation, and review.

Main Street Wealth Open Source

A collection of valuation and due diligence tools for home services mergers and acquisitions.

A collection of tools and resources to help people reviewing small and midsize home services deals examine business value, earnings metrics, customer concentration, and deal structure.

whats-agent-doing

What's Agent Doing: a Claude Code plugin with a little box above the prompt that always says what Claude is doing right now - tzafrir/whats-agent-doing

outis

A tool that sends fake address-not-found bounce notices in response to received spam.

Helps users who want to reduce automated email create a 'user unknown' bounce notice from a received message and send it to the sender.

snip

A tool for searching terminal notes and letting Claude read and write them.

Provides SQLite full-text search for people managing notes in a terminal, and lets Claude save, search, and edit the same notes through an MCP server.

jaynshare

A self-hosted proxy for choosing which connected account to use in a Claude Code session.

Lets users who want to connect multiple Claude accounts register them on a server, connect clients, and choose an account for each session.

hn_whos_hiring

Rank every HN 'Who is hiring?' post against your resume, then draft emails and prefill applications from a local dashboard.

libredb-studio

One browser tab for PostgreSQL, MySQL, Oracle, SQL Server, MongoDB, Redis, SQLite, Couchbase, ClickHouse, Druid, DuckDB, Turso and more.

ldraw-nova

Agent tooling for generative LEGO models building, built with Astra and Opus 5.5, powered by Jev - anteloc/ldraw-nova

system

An open-source system for apps built with web technology.

Audionaut

An audio editor for multitrack recording and editing that can also delegate work to AI agents.

A free, open-source desktop app for people who edit multiple audio tracks for music and podcasts. It offers per-track playlists and precise trimming, and lets AI agents edit sessions through MCP.

gridpath

A tool that lets AI edit actual Excel files and users review changes before saving.

A tool for users who want to modify existing .xlsx files with AI while checking the changes. They can edit files through an MCP server or desktop app, review cell-by-cell before-and-after differences, and then save.