Today's drop: 2 new tools, 2 listing updates
The directory now lists 145 hand-checked tools. New today: Stroq, a free local CLI that reads your past coding-agent sessions to show which keys left your machine and then guards the next session, and Orca, the open-source desktop app for running Claude Code, Codex and other agents in parallel, each in its own worktree. Updated: CISO Assistant (v4.2.2) and Strands Box (Linux ARM64 build). AI for IT & Cybersecurity grows to 56 tools and Coding & Vibe Coding to 25. Pricing only where the official page states it.
Previous drop: Nov 7, 2026 →
Get the weekly dropThe best new AI tools, one email a week. Free.
New in the directory 2
Browse all 145Tool
Category
Price
Status
Link
StroqFree, open-source (Apache-2.0) local security CLI for AI coding agents. Run npx @stroq/cli and it reads your last 30 days of Claude Code and Codex sessions and shows how many commands came from text the agent read and which keys left your machine; stroq sent checks Claude Code, Codex CLI and Cursor sessions against the credential files on your computer and reports where each key went, printing key names, never values. stroq init installs hooks (Claude Code, Cursor, Codex, Copilot CLI, Windsurf, Antigravity, plus an MCP proxy) that deny curl-pipe-to-shell and outbound calls carrying a known secret, and ask you first when a command repeats text the agent read. A local console on 127.0.0.1 shows every decision with its reason. No account, no telemetry; needs Node.js 22+. Matching is text-level, so it is a filter, not a sandbox. About 210 GitHub stars and about 600 npm downloads in the week of Nov 1-7, 2026; v0.30.0 shipped Nov 5 alongside a Show HN post.
Price Free and open source (Apache-2.0)
OrcaFree, open-source (MIT) desktop app from Stably AI (Y Combinator) for running several coding agents in parallel, each in its own git worktree. Bring the CLI agents and subscriptions you already have (Claude Code, Codex, Gemini, GitHub Copilot, OpenCode, Cline, Grok and any other terminal agent), use each in a chat view or the raw terminal, and review diffs, leave line notes and send PR review comments or failing checks back to an agent. A coordinator agent can split a big task across worker agents; workspaces can run locally, on an SSH host or a cloud VM, and an iOS and Android companion app pings you when an agent needs input. Prompts go straight to the providers you configure. Builds for macOS, Windows and Linux. About 119,000 GitHub stars; ships almost daily (v1.4.252 on Nov 6, 2026 added chat forking and Plan mode).
Price Free and open source (MIT); Enterprise rollout by contact
Why these two
- Stroq answers a question most guardrail tools skip: what did my agent already do? Run
npx @stroq/cliand it reads your last 30 days of Claude Code and Codex sessions;stroq sentcompares Claude Code, Codex CLI and Cursor sessions with the credential files on your machine and says where each key went, printing names, never values.stroq initthen adds hooks that denycurl | shand outbound calls carrying a known secret. Free (Apache-2.0), local, no account or telemetry. About 210 GitHub stars, about 600 npm downloads last week and a Show HN on Nov 5 with v0.30.0. Its own README is clear about limits: text-level matching, a filter rather than a sandbox. It pairs with HOL Guard and Failproof AI on the guard side, and with Strands Box when you want a real sandbox. - Orca is a free, MIT-licensed desktop app from Stably AI with about 119,000 GitHub stars. It does not ship its own model: you plug in the agent CLIs and subscriptions you already pay for (Claude Code, Codex, Gemini, Copilot, OpenCode, Cline and others) and run them side by side in separate worktrees, with diff review, line notes, a coordinator that splits work across worker agents, remote and cloud workspaces and a phone app that pings you when an agent needs input. It is free, ships almost daily, and is now on our Cursor alternatives page.
Listing updates
Checked on Nov 8, 2026 against official pages and release notes.
- CISO Assistant: v4.2.2 (GitHub release Nov 7, 7:56 PM ET) adds MCP tools that let an AI assistant assess and write findings, an assets lifecycle, a cumulative outcome for Australia's Essential Eight, new framework libraries (ASD ISM September 2026, ASD Essential Eight October 2026, Swift CSCF 2026 and 2027, SAMA BCM, Germany's 2025 NIS-2 implementation act), a better ISO 27001 to NIST CSF 2.0 mapping and security hardening.
- Strands Box: v0.2.0 and v0.2.1 (Nov 5). v0.2.1 adds a Linux ARM64 build (glibc 2.34 or newer, with steps for Amazon Linux 2023) next to the Apple silicon Mac build; the release notes say package availability does not establish security acceptance. About 500 GitHub stars.
Category changes
- AI for IT & Cybersecurity grows from 55 to 56 tools with Stroq.
- Coding & Vibe Coding grows from 24 to 25 tools with Orca.
Checked, not listed
- Observer (~1,700 stars, AGPL-3.0): local agent that watches a camera, screen or mic with local models; interesting, but outside our coding and IT focus for now.
- Terrarium (~56 stars): local microVM sandbox for agents; too early, overlaps Strands Box.
- Valet (~27 stars): AI homelab dashboard; too early.
- leeway and TraceMind (~1 star each): too early.
- Obsessive-1.7B: a single fine-tuned model release on Hugging Face; we list tools rather than individual fine-tunes.
- claude-mem and Vercel agent-desktop: large and active; on our watch list for a future agents and memory roundup.