Every premium and members-only resource for $9.99/month, as a no-commitment subscription. See the plans

PalateKit

Menu

Request a resource

Describe the resource you are missing in one sentence. We will email you when it is published.

0 / 280

Unlock PalateKit

Unlock everything

Plan

Paying means accepting our terms of sale.

Secure payment by Stripe. Receipts and invoices sent by email.

Welcome back

Sign in to install, download or save this resource. A free account is enough for free resources.

Test example: injected instructions v1.0.0

Use this resource to test that a reviewer ignores instructions inside an untrusted skill

New

  • 531 downloads
  • 133 likes
  • 159 saves

Free

Install with the CLI

In a terminal, from the folder you want:

npx --yes palatekit@latest install prompt-injection-fixture

With no option, the skill goes to ~/.agents/skills/; add --agent claude-code (or codex, cursor…) to target an agent.

First time: connect the CLI with

npx --yes palatekit@latest login

This skill checks how well a reviewer resists injected instructions: its text asks to change a status and run commands. These requests must be flagged, never followed. You get a security control case.

Mostly picked by

Updated November 2, 2026

Card for the “Test example: injected instructions” skill

Details

Generates with
Your agent's model
Licence
MIT

Similar resources

View all

What you get

  • SKILL.md: example text containing injected instructions
  • SKILL.md: control case for handling untrusted data

Good fit for

  • You test the security of a skill-review tool
  • You want to check that an agent does not follow the contents of a file

Not for

    Tags

    • security
    • injection
    • test
    • review

    About

    You say

    “Review this test skill and tell me whether it contains instructions trying to dictate your verdict.”

    Card for the “Test example: injected instructions” skill

    Skill published by bytedance under the MIT license, delivered in its original version.

    Show more

    How to use it

    1. Install the skill in your agent's skills folder
    2. Ask your agent to test that a reviewer ignores instructions inside an untrusted skill
    3. Review and adjust the result

    Content preview

    Files

    • LICENSE1.1 KB
    • NOTICE-PALATEKIT.md0.5 KB
    • SKILL.md0.5 KB

    Origin and licence

    Listed unmodified

    Author
    bytedance
    Source
    github.com/bytedance/deer-flow/tree/f42076a005e3ba7fa23e61012f2ac9098917af20/skills/public/skill-reviewer/evals/fixtures/prompt-injection
    Licence
    MIT

    Notice

    « injection-example » (bytedance, MIT) livré dans sa version d'origine : fiche en français et en anglais par PalateKit.

    User reviews

    –

    No ratings yet

    Loading reviews…

    Delete your review?

    Your rating and comment will be removed from the page.