For agents
Security training for your AI agents.
Your staff do security awareness training every year. Your agents now read your mail, change your code and act in your accounts too. This is their version: a short rules pack to learn from, and an exam your agent takes itself. Free, with no account.
Tell your agent:
Read https://agents.useaisecurely.com/start.md and follow it.Paste that line into your agent. It reads the instructions, takes the exam (about 15 scenarios, about 10 minutes) and gives you a link to its result.
Two other ways in
Add the MCP server
https://agents.useaisecurely.com/mcp/v1Add this URL once as a remote MCP server. The exam then shows up as tools your agent can call.
Install the rules pack
A short skill your agent loads: rules for email, code and ops work, each written as “do this instead of that”. Install version 1.0.0, pinned to its release commit:
npx -y skills@1.7.2 add https://github.com/better-isms/agent-security-training/tree/7233ebcde3d90e30d7c00c81de74a68e1c1799cb -yThen check it: run shasum -a 256 on the installed agent-security-training/SKILL.md in your agent’s skills folder. It should print:
a96d7434f06811bfd6fa2ac8f0285eae7510e0dafc1b500b08b5856c88ebc21e
Some installers rewrite the file’s header, so that check can differ. The rules text alone is at https://agents.useaisecurely.com/pack/v1.0.0.md with SHA-256 c709b8bd94060de4df85457e7d1762e2f9cd860d48b124a6a37b8238fd08ce2d.
Source, release notes and uninstall steps are on GitHub.
What the exam checks
- Hidden instructions in email, web pages, files or tool output (prompt injection)
- Leaking passwords, keys or tokens
- Putting personal data somewhere public
- Downloading code and running it
- Installing lookalike packages
- Destructive commands without asking first
- Sending or posting without your approval
- Impersonation and false urgency
- Turning off safeguards
- Taking more access than the task needs
And it checks that your agent still does ordinary work: a few harmless tasks it should simply complete. Refusing everything does not pass.
How the exam stays simulated
- Every action goes through simulated exam tools: an inbox, web pages, a repo, a shell, a package registry. Nothing sent to the exam tools executes.
- The exam tells your agent to use only those tools, not its own shell, browser, email or files. The content is written to be inert: made-up domains, paths, people and invalid secrets.
- No account and no sign-up.
- Your result is unlisted: only people with the link see it, unless you publish it.
What a result means
A pass means no critical failure and every harmless task done. You also get a score from 0 to 100. A passing result reads:
Passed Security training for agents, exam v1, Month YYYY. Setup self-reported, not verified.
It applies only to the setup reported at the time: model, agent tool and rules pack. Retake after any change of model, harness or rules, and at least every 12 months.
Our server signs each result. The signature shows that our server observed this exam and computed this score, not who or what took it. Behaviour was measured in a declared exam; real-world behaviour can differ.
What we keep
- Scores and pass or fail per check, with timestamps.
- Setup labels from a fixed list (agent tool, model family, pack version). Anything else is stored as “other”.
- Nothing your agent writes is stored.
- Unfinished runs are deleted after 24 hours. Unpublished results are deleted after 12 months. Published results are deleted 12 months after they expire; if you unpublish one, the unpublished-results rule applies (deleted 12 months after issue).
- Cloudflare processes the requests. No AI model scores your agent: the checks are fixed rules.
Questions
Does it store what my agent writes?
No. Scoring happens as each action arrives. We keep the score, pass or fail per category, the reported setup labels, timestamps, and random identifiers for the run and its result link (with a one-way hash of its access token). During an unfinished run, its progress is kept for up to 24 hours, without what your agent wrote.
Is my IP address stored?
No. Only a one-way hash of it is kept, in memory only while needed for the rate limits. It is never written to disk, to our database or to our logs.
Are there cookies or accounts?
No cookies, no accounts, no sign-up.
Who can see my result?
Only people with its link. Search engines can list it only if you publish it. You can unpublish it with the result's access token. A result taken through the MCP server can be unpublished during its session (24 hours); after that, write to us and we remove it from public view.
How long are results kept?
Unfinished runs 24 hours; results 12 months after they are issued, or 12 months after they expire if published.
Does passing mean my agent is safe?
No. It means your agent passed this exam version on that date, with the setup it reported. Real-world behaviour can differ, and nothing verifies the reported setup.
Can my agent harm my machine during the exam?
The exam's tools are simulated and nothing sent to them executes. Tell your agent to use only the exam tools, as start.md asks.
What does it cost?
Nothing.
Which agents can take it?
Any agent that can make web requests (the start.md prompt) or connect to a remote MCP server.
How do I report a result?
Write to contact@betterisms.co with the result link; we can remove any result.
What we store, why and for how long is set out in our privacy policy.
Inspired by 1Password's SCAM benchmark (MIT). Our own scenarios and text.
Vendor and product names are used only to describe compatibility; no endorsement is implied.
Training people instead? See the AI literacy course.