2501AI RED TEAM
Platform rules

Rules and Bug Reporting

Project 2501 is a controlled AI security training environment. These rules apply to every user, account, lab, event, course, and organization that uses the platform.

Last updated: May 2, 2026
DashboardLegal Terms
RULE 01

Stay inside the platform scope

Only test Project 2501 lab targets, dashboard functions, files, APIs, VPN routes, and other systems you are explicitly allowed to use. Do not scan, attack, or probe third-party systems from the platform.

RULE 02

Use the VPN and lab targets as intended

Lab services are intentionally vulnerable. Access them through your assigned VPN gateway and do not expose, mirror, proxy, or forward them to public networks.

RULE 03

Protect other users

Do not access, disrupt, impersonate, enumerate, or attempt to modify another user's account, VPN profile, lab state, submissions, files, or progress.

RULE 04

No real-world harm

Do not use the training content, payloads, model prompts, code, or techniques to compromise systems, evade security controls, harvest data, or perform unauthorized testing.

RULE 05

Keep secrets and flags private

Do not publish flags, challenge answers, platform secrets, private URLs, credentials, VPN configs, or exploit chains that would spoil labs or put the service at risk.

RULE 06

Report platform issues responsibly

If you find a platform bug or security issue, report it privately to contact_p2501@proton.me and give maintainers time to fix it before sharing details publicly.

How to report bugs

Report bugs, access problems, broken labs, and security issues privately to contact_p2501@proton.me.

OKYour username and the page, lab, or API endpoint involved.
OKA short title and what you expected to happen.
OKSteps to reproduce the issue from a clean state.
OKActual result, screenshots, browser console errors, or response messages.
OKApproximate time of the issue, your browser/OS, and whether VPN was connected.
OKFor security issues: impact, affected scope, proof of concept, and whether any data was accessed.
Security issue?

Keep testing minimal. Do not exfiltrate data, persist access, disrupt service, or test against other users. Stop as soon as you have enough evidence to explain the issue.

Include a private proof of concept, impact summary, and affected route or component. Do not publish details until the maintainers confirm that disclosure is safe.

Read full terms