DAILY DROP · NOV 1, 2026

Today's drop: 2 new tools, 2 listing updates

The directory now lists 131 hand-checked tools. New today, both for IT and security: AiSOC, a free, self-hosted AI security operations center that triages alerts with a local model and logs every step; and RedAmon, an open-source AI pentesting framework with human approval gates that can open pull requests to fix what it finds. Updated: Claude Code (a setting that blocks writes outside your project) and Perplexity (Computer now works in your own Chrome). AI for IT & Cybersecurity grows to 47 tools. Pricing only where the official page states it.

Previous drop: Oct 31, 2026 →

Get the weekly dropThe best new AI tools, one email a week. Free.

New in the directory 2

Browse all 131
Tool
Category
Price
Status
Link
AiSOCFree, open-source (MIT), self-hosted AI Security Operations Center. It ingests telemetry from your security tools (87 connectors including Splunk, Sentinel, Elastic, CrowdStrike, Okta, AWS and Kubernetes audit, plus a syslog/CEF listener), runs detection rules, groups alerts into incidents and has an AI agent triage each one, recording every prompt and tool call in a replayable Investigation Ledger. A human approves any response before it reaches a vendor. It ships a small local model (Llama 3.2 3B via Ollama) so it runs on CPU with no API key, plus an MCP server for Claude, Cursor and Continue and a translator for Splunk SPL, Sentinel KQL and Elastic EQL rules. Needs Docker Compose with 8 GB memory and 20 GB disk. About 5,800 GitHub stars; v21.0.0 shipped Nov 1, 2026. The project publishes its own audit of what does and doesn't work yet.
Price Free and open source (MIT)
FREE TIERchecked 11/01
RedAmonFree, open-source (MIT), self-hosted AI penetration testing framework. It runs recon tools in parallel and maps your attack surface into a Neo4j graph, then an AI agent tries to exploit what it found from a Kali sandbox, with human approval gates before risky steps, and a CodeFix agent can open GitHub pull requests that fix the issues. Works with MCP in both directions (plug in MCP servers as tools, or drive RedAmon from Claude Code). Runs in Docker; needs about 7 GB RAM and 80 GB disk without OpenVAS. The project reports solving 101 of 104 XBOW web-security benchmark challenges black-box; that is its own claim. About 3,300 GitHub stars. For systems you own or have written permission to test only.
Price Free and open source (MIT)
FREE TIERchecked 11/01

Why these two

  • AiSOC is a free, MIT-licensed AI security operations center you run on your own server. It pulls in telemetry from your security tools (87 connectors, including Splunk, Sentinel, Elastic, CrowdStrike, Okta and AWS), runs detection rules, groups alerts into incidents and has an AI agent triage each one. Every prompt, tool call and citation goes into a replayable Investigation Ledger, so you can check how the AI reached a verdict instead of trusting a summary, and a human approves any action before it reaches a vendor. It ships a small local model (Llama 3.2 3B through Ollama) that runs on CPU with no API key, plus an MCP server for Claude, Cursor and Continue. It has about 5,800 GitHub stars and shipped v21.0.0 today. One caution: the bundled model is small, so treat its verdicts as a first pass; the project itself publishes an audit of what works and what doesn't yet.
  • RedAmon is an open-source (MIT), self-hosted AI penetration testing framework. It runs recon tools in parallel, builds a graph of your attack surface, lets an AI agent attempt exploits from a Kali sandbox with human approval gates and can open GitHub pull requests that fix what it found. It speaks MCP both ways, so you can drive it from Claude Code. The project says it solved 101 of 104 XBOW web-security benchmark challenges black-box; that is its own claim. About 3,300 GitHub stars. It needs a fairly big machine (about 7 GB RAM and 80 GB disk without OpenVAS) and is for systems you own or have written permission to test only.

Listing updates

Checked on Nov 1, 2026 against official pages and announcements.

  • Claude Code: v2.1.314 (Oct 30) adds permissions.blockWritesOutsideWorkingDirectories. Turn it on and Write, Edit and NotebookEdit refuse any path outside your working directories, in every permission mode, which is a simple guardrail if you let the agent run with fewer prompts. v2.1.316 (Oct 31) adds CLAUDE_CODE_DISABLE_ULTRATHINK and CLAUDE_CODE_SUBAGENT_EFFORT.
  • Perplexity: on Oct 30 Perplexity said a new Chrome extension lets Computer in its Mac and Windows apps work inside the Chrome you already use, so it can use sites you're signed into. The extension listing says Edge, Brave, Arc and other Chromium browsers work too, and that Computer uses the browser only during a task you start. Comet doesn't need it.

Category changes

  • AI for IT & Cybersecurity grows from 45 to 47 tools. The Breaking into IT guide adds AiSOC (run it in your home lab with sample data and review each AI triage verdict like a Tier 1 analyst) and RedAmon (point it only at a deliberately vulnerable practice app in an isolated lab).

Checked, not listed

  • Vigil SOC 1.0 (Oct 30): another open-source AI SOC that hit general availability, with about 400 GitHub stars. Worth watching, but AiSOC has far more users today.
  • Pipelock: an open-source agent firewall that scans an agent's HTTP and MCP traffic for leaked secrets and prompt injection (about 950 stars). Solid, but it overlaps HOL Guard and Project Quarry, which we already list.
  • Mitigant MCP server (Oct 31): lets agents run Mitigant's cloud attack emulation, but it is for existing Mitigant customers.