PromptShieldCatch prompt injection before customers do.
Paste untrusted samples against your app description. Get deterministic pattern hits plus a hardening checklist — not a penetration-test certificate.
No credit card required · Cancel anytime
Key Takeaways
- Deterministic injection / jailbreak pattern checklist.
- Fired patterns called out with hardening guidance.
- For LLM/agent product + security teams — decision-support, not a pen-test cert.
Untrusted input. Injection patterns that fire.
Decision-support scans — not a certified security assessment.
PromptShield — definition
PromptShield scans an AI app for prompt-injection and jailbreak risk before you ship — paste untrusted input samples, get the patterns that fired, and a hardening checklist for LLM/agent product teams.
What this is NOT
- Not a penetration-test firm, red-team retainer, or ISO certification body.
- Not a guarantee that a clean scan means the model is safe in production.
- Not a penetration-test certificate — have security review high-stakes launches.
Controller / Processor — When packs mention privacy roles, we use GDPR Art.4 labels: Controller and Processor (decision-support only).
Customer input is untrusted. Jailbreaks ship fast.
Prompt injection
Untrusted text overrides system instructions.
Jailbreak overrides
Role-play and delimiter tricks bypass safety rails.
Tool abuse
Agents with tools amplify a single successful injection.
Late discovery
Issues surface in production abuse, not in happy-path demos.
Describe. Paste samples. Harden.
Deterministic pattern hits plus a hardening checklist.
Pattern checklist
Injection and jailbreak-oriented signals against your app description.
Fired patterns called out
See what matched — not a opaque dump.
Hardening guidance
Checklist items for LLM/agent product and security teams.
Honest demo vs live
Demo is labelled; live narrative needs a configured key.
Everything you need to scan injection risk early
Built for LLM/agent teams shipping customer-facing chat with untrusted input.
Deterministic injection pattern checklist
Injection and jailbreak-oriented signals against your app description.
Injection pattern detection
See what matched — not a opaque dump.
Jailbreak signal flags
Checklist items for LLM/agent product and security teams.
Hardening checklist
Demo is labelled; live narrative needs a configured key.
From sample to hardening checklist in 3 steps
Describe
Describe the AI app and paste untrusted samples.
Scan
Deterministic patterns plus optional model narrative.
Harden
Apply checklist items; keep humans for security review.
Describe. Paste samples. Harden.
Deterministic pattern hits plus a hardening checklist.
Prompt injection
Untrusted text overrides system instructions.
Pattern checklist — Injection and jailbreak-oriented signals against your app description.
Try it on this page
or run an injection scan below.
No testimonials yet, by design
We don't publish customer counts or invented quotes. Run a scan and read the fired patterns — that's our proof.
Injection pattern hits
Paste an app description and untrusted samples. See which injection and jailbreak patterns fire. Try a labelled demo without a paid plan.
Hardening checklist
Each scan lists concrete hardening steps for the boundary between user text and your system prompt. You still own the fix.
Honest scan limits
Free 3/day · 15/mo; Pro 40/day · 300/mo. No hidden caps. Demo mode is labelled. Not a pentest or a compliance certificate.
Honest boundaries
PromptShield is decision-support for injection risk — not a penetration-test certificate.
Simple plans that scale
All paid plans include
- Injection scan studio
- Deterministic pattern checklist
- Hardening guidance (decision-support)
- Email support on paid plans — no hidden add-ons for core studio use
- 3 injection scans/day · 15/mo
- Studio demo (no extra LLM fee)
- AI included: 40 scans/day · 300/mo · fair use
- No separate ChatGPT subscription required
- Priority support · 2 months free with yearly
Feature comparison
What each plan includes — quotas reflect the enforced fair-use limits.
| Dimension | Free | Pro | Enterprise |
|---|---|---|---|
| Usage | |||
| AI runs | 3/day · 15/mo | 40/day · 300/mo | 200/day · 3,000/mo |
| Report export | Watermarked | Clean export | Clean export |
| Audit log | Not included | ✓ Included | ✓ Included |
| Product | |||
| Deterministic injection pattern checklist | ✓ Included | ✓ Included | ✓ Included |
| Injection pattern detection | ✓ Included | ✓ Included | ✓ Included |
| Jailbreak signal flags | ✓ Included | ✓ Included | ✓ Included |
| Hardening checklist | ✓ Included | ✓ Included | ✓ Included |
| BYOK (your LLM key) | Not included | Not included | ✓ |
| SSO | Not included | Not included | Roadmap |
| Support & compliance | |||
| Support | Community | Priority | Custom SLA |
Already paid? Unlock Pro with your license
Licenses are issued manually after your payment is confirmed: we generate one per paid order and send it to the purchase email, typically within a few hours. Nothing yet? Write to lixingliangsy@163.com and reference your order email — we will help activate Pro on this device.
Your questions, answered
What does PromptShield scan?
Prompt-injection and jailbreak risk signals in untrusted input against your app description — with a hardening checklist. Decision-support, not a penetration-test certificate.
Is this legal advice or a certification?
No. Outputs are decision-support drafts. Have counsel and risk owners review before production. We do not claim ISO, CMP, or regulator certification.
Can I cancel anytime?
Yes. Self-serve plans cancel anytime; access continues until period end.
Do I need a credit card to start?
No. Start with email signup or Demo mode in the studio, then upgrade when ready.
Do I need my own ChatGPT / OpenAI subscription?
No for Free/Pro. AI runs are included in your plan (fair use) via our platform key. Enterprise can optionally bring your own OpenAI key (BYOK) — configure it at /settings (key stays server-side only).
What is Fair Use / what if I hit the AI limit?
Free is 3 scans/day and 15/month. Pro is 40 scans/day and 300/month. If you hit the fair-use limit, Studio returns a labelled demo result until the daily or monthly reset — or upgrade / use Enterprise BYOK for heavier volume.
Is checkout secure?
Payments are processed by Waffo Pancake (merchant of record).
What happens after I pay?
You receive access confirmation; fulfillment is tracked via webhook + order logs.
Quick answers
- Deterministic injection pattern checklist
- Injection pattern detection
- Jailbreak signal flags
- Hardening checklist
- CI-ready (Pro)
More questions
Which threats?
Prompt injection, jailbreak-style overrides, and common untrusted-input patterns (OWASP LLM01-oriented).
Is it a pen test?
No. Deterministic pattern checklist + optional model narrative — not a certified security assessment.
What inputs?
App description plus sample untrusted user/tool input.
Demo vs live?
Demo is labelled; live AI narrative needs a configured key.
Who is it for?
LLM/agent product and security teams shipping customer-facing chat.
Primary sources?
OWASP LLM Top 10 (LLM01 Prompt Injection) and related agent threat notes.
How it compares
| Dimension | Manual | PromptShield |
|---|---|---|
| Speed | Hours to days | Minutes per run |
| Consistency | Varies by person | Same ruleset every run |
| Output | Free-form | Structured, exportable result |
| Best for | Final sign-off | First-pass decision-support |
When not to use: Use qualified human review instead — it is a decision-support risk pass, not a formal pentest or compliance attestation.
- promptshield howto 2026 — From the PromptShield blog.
- promptshield overview 2026 — From the PromptShield blog.
- Fleet launch notes — PromptShield ships in the LX AI Micro-SaaS fleet.
People also search
Click a tab to expand it; click again to hide. All content stays on the page for search engines.
Start smarter today
Join builders using PromptShield. Free to try — no card needed.
Trusted · Cancel anytime