Prompt Fuzzer

Open-source tool that runs dynamic attacks against a GenAI app's system prompt to help harden it.

Observability tool by Prompt Security (SentinelOne)

IndexedMIT★ 719

Install

Pick your client. Read the repository before you give any tool access to your data.

pip install prompt-security-fuzzer

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

About

Prompt Fuzzer is an interactive tool that assesses a GenAI application's system prompt by simulating LLM-based attacks, then reports a security evaluation of the results. It includes a playground for revising the prompt and retesting it. The README lists 16 LLM providers and 16 attack types, installs with pip, and warns that runs consume tokens.

Use cases: security-scanning, prompt-engineering

Reviews

Write a review

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Fact sheet

Category: Red teaming · open source
Open source: Yesraw.githubusercontent.com

Facts read from the vendor's own pages by claude-haiku-5-5 on Oct 24, 2026; a fact is shown only with the page it came from.

Signals

Stars: 719
Forks: 103
Open issues: 16
Last commit: Aug 19, 2026 (2 months ago)
Last release: Aug 18, 2026 (2 months ago)

From GitHub (fetched Oct 24, 2026).

Details

Packages: pypi prompt-security-fuzzer
License: MIT
Repository: github.com/prompt-security/ps-fuzz
Website: github.com/prompt-security/ps-fuzz
First seen: Oct 24, 2026
Last checked: Oct 24, 2026
Source: README github-graphql

Cite this page

Prompt Fuzzer. Agent Mart. https://agentmart.dev/observability/prompt-fuzzer (last updated 2026-10-24)