ForkFree

Security and privacy

Stroq

by AGGIB

Stroq is a command-line tool that scans your past Claude Code, Codex and Cursor sessions to show which API keys were sent elsewhere and which commands came from text the agent had read. It can also install local hooks that block or question risky commands before the agent runs them.

Proof level: ReadWorth a lookApache-2.0

Get it on GitHubProject websiteCompare

No demo yet

Proof level: Read. We read its README and licence and ran our automatic checks. We have not run it yet. So this page is based on its README and our automatic checks.

The catch

Worth a lookEstablished, with a few things to check (listed below).

Check this first: Findings are based on today's policy and on matching keys found in commands, so it may miss some leaks or flag false ones.

Things to check

  • Findings are based on today's policy and on matching keys found in commands, so it may miss some leaks or flag false ones
  • Relatively young project with a modest following
  • Adding hooks from the console works only on macOS and Linux

Can you use Stroq commercially?

Free to use, change and sell, including in paid products. Keep the copyright notice.

Not legal advice. Read the project's licence file before building a product on it.

What can you do with Stroq?

  • Scan the last 30 days of agent sessions to see which keys were sent to other companies, so you know what to rotate
  • Install hooks that deny curl-to-shell commands and outbound calls carrying known secrets, and ask before risky commands run
  • Inspect a folder you were sent before running it, and use the local console to review decisions, skills and MCP servers

How do you get Stroq running?

Run it with npx @stroq/cli, which needs Node.js 22 or newer; no account is needed.

Follow the official instructions in the project README.

Make money with it

A freelancer could offer an AI-agent security check for small teams, using the session scan and hook setup as part of a paid report.

An idea, not financial advice. Check the licence before charging for anything built on it.

Ways a business could use it

  • Check whether client or shop credentials, such as payment or cloud keys, leaked through AI agent sessions, so you can rotate them before they cause losses
  • Put guard hooks on the coding agents used by a small team to reduce the risk of prompt-injection attacks leaking data or running malicious commands

Who's talking about it

Checked 08 Nov 2026. Summary drafted by AI from the project's README. Proof level 1 of 4, Read: we read its readme and licence and ran our automatic checks. we have not run it yet.

Similar tools in Security and privacy

sherlock sherlock-project
Worth a look

Sherlock is a command-line tool that checks whether a given username exists on many social networks and websites. It is used to find a person's or brand's online accounts for research and investigation.

Proof level: Read 114.8k stars +794 a day MIT
ARTEX BaaRai
New and promising

ARTEX is a self-hosted system that uses an AI model to carry out penetration testing of websites and networks on its own, with a web dashboard to track tasks, findings and assets. It is meant for security testers who want to automate exploration of targets they are authorised to test.

Proof level: Read 12.6k stars +538 a day AGPL-3.0
bla qybaihe
New and promising

BLA is a command-line tool and MCP server that hardens software releases such as Electron apps, JavaScript bundles, WebAssembly and native binaries against AI-driven reverse-engineering agents. It adds decoys and resource-exhausting structures so automated analysis becomes slow, costly or inaccurate.

Proof level: Read 6.2k stars +258 a day MIT