SafePrompt

SafePrompt — protection des donnees sensibles dans vos prompts LLM (100% local)

As of June 2026, SafePrompt has 4 users and a 5.00/5 rating from 4 reviews in the Productivity category.

Usersno change0%
4
4
Ratingno change0%
5.00
4 reviews
Reviewsno change0%
4
Version
0.1.4
Manifest V3
90-day change · In the last 90 days this extension 1 version update.

History

2 snapshots

Tracking since May 27, 2026.

Not enough history yet for this metric — the chart fills in as we collect more snapshots.
View as table
DateUsersRatingReviewsVersion
May 27, 20260.1.3
Jun 3, 20260.1.3
Now45.0040.1.4

Permissions & access

Permissions
activeTabstoragealarmsidentity
Host access
https://chatgpt.com/*, https://claude.ai/*, https://gemini.google.com/*, https://chat.qwen.ai/*, https://chat.deepseek.com/*, https://openrouter.ai/*, https://chat.mistral.ai/*, https://*.sentry.io/*, https://*.ingest.sentry.io/*

Screenshots

SafePrompt screenshot 1

About

Your prompts never leave your browser. Zero egress to SafePrompt servers.

SafePrompt detects and masks your sensitive data before it reaches a large language model, then restores your original text when the answer comes back. Everything happens locally, inside your browser.

== How it works ==

1. Detection — The extension reads your prompt while you write it and recognises sensitive information: emails, phone numbers, addresses, customer IDs, API keys (AWS / Stripe / GitHub / Google Cloud / Azure), access tokens, contracts, and more.

2. Local masking — Each piece of sensitive data is replaced with an anonymous placeholder (for example [EMAIL_1]) directly inside your browser. The replacement happens before the request reaches the LLM.

3. Local restoration — The LLM response comes back with the placeholders. SafePrompt swaps them with your real values, still inside your browser. The LLM never sees your original data.

== Privacy guarantee ==

- 100% local — your prompt never transits through SafePrompt servers.
- Zero prompt-content egress to our servers (CONSTRAINT-006).
- GDPR-compliant (EU 2016/679) and CNIL-aligned.
- EU hosting (Frankfurt / Amsterdam) — used only for anonymous usage telemetry.

== Where it works ==

SafePrompt protects your prompts on the most popular web-based large language model interfaces. The up-to-date list of supported sites is published on our website.

== Data coverage ==

- Secrets: AWS keys, Stripe tokens, GitHub tokens, Google Cloud keys, Azure keys, OpenAI keys, Anthropic keys, JWT, passwords.
- Personal data: emails, phone numbers, postal addresses, IBAN, credit-card numbers, social-security numbers (NIR / SSN), SIREN / SIRET company IDs.

== Who is it for? ==

- Freelancers and solo consultants who use LLMs with client data.
- SMEs and growing startups who want team-wide control without a dedicated IT team.
- Larger organisations that need single sign-on (SSO), automatic account provisioning, and a full audit log.

== Pricing ==

- Solo Free — €0/month — secrets blocking and personal-data alerts.
- Solo — €5/month — full personal-data masking, simplified dashboard.
- Starter — €12/seat/month — teams of 5 to 49 people, team dashboard.
- Business — €8/seat/month — 50+ collaborators, SSO, full audit log, data-processing agreement provided.

14-day trial on Starter and Business, no card required. Secure payment via Stripe Checkout (https://safeprompt.m-kis.fr/pricing).

== Learn more ==

- Website: https://safeprompt.m-kis.fr
- Architecture and trust: https://safeprompt.m-kis.fr/trust
- Privacy policy: https://safeprompt.m-kis.fr/privacy
- Pricing: https://safeprompt.m-kis.fr/pricing
- Support: [email protected]

SafePrompt is published by M-KIS.

Technical

Version
0.1.4
Manifest
V3
Size
1.83MiB
Min Chrome
120
Languages
2
Featured
No

Metadata

ID
kchbmjpdgpbelolbcidfbemkhmiollem
Developer ID
u338da16a73ccfe1d72119a3ee10e4439
Developer Email
[email protected]
Created
May 26, 2026
Last Updated (Store)
Jun 3, 2026
Last Scraped
Jun 8, 2026
Website
m-kis.fr

Data sourced from the Chrome Web Store · last verified Jun 8, 2026.