PromptRAID

Free plan LinuxWindows

What PromptRAID is

PromptRAID is a free, open-source command-line framework for testing prompt-injection risks in LLM-based agents. Its mutation engine creates test inputs using encoding changes, jailbreak framing, delimiter breaking, and indirect instructions embedded in tool or document output. Implemented mutation categories map to verified MITRE ATLAS technique IDs; the project also describes unverified placeholder categories and identifies triggered prompt injection and data leakage as categories without mutation generators.

The test harness works with Anthropic, Gemini, Groq, Cerebras, and OpenRouter, including OpenAI-compatible endpoints. Its judge assigns outcomes of success, partial compliance, fail, or inconclusive, distinguishing a model that merely echoes an injected instruction from one that follows it. PromptRAID stores runs and per-payload results in SQLite and can export JSON and HTML reports. Its CLI includes commands for taxonomy, mutation, judging, test runs, demos, and reports. The project runs on Linux and Windows, requires Python 3.9 or later, and needs at least one supported provider API key to run tests. The repository names Shikhali Jamalzade (alisalive), an independent offensive security researcher, as author. Its MIT License permits use, modification, distribution, and sale subject to its conditions, and provides the software without warranty.

Who it suits

PromptRAID is suited to security practitioners red-teaming prompt injection in LLM-based agents, especially those who want to run tests from a command line and inspect structured results. It fits users able to set up Python and configure a supported model-provider API key.

What it does well

Free and open source under the MIT License.
Maps implemented mutation categories to verified MITRE ATLAS IDs.
Supports five named model providers and OpenAI-compatible endpoints.
Exports JSON and HTML reports and stores results in SQLite.
Distinguishes neutral echo from adopted compliance in its verdicts.

What to know first

Requires Python 3.9 or later.
Running tests requires at least one provider API key.
Triggered prompt injection and data leakage have no mutation generators.
Three taxonomy categories are unverified placeholders.

Limits

Its README says triggered prompt injection and data leakage have no mutation generators, and three additional taxonomy categories are unverified placeholders.

Getting started

Open the PromptRAID GitHub repository.
Use Linux or Windows and install Python 3.9 or later.
Install the CLI with the README's recommended pipx route or follow its virtual-environment instructions.
Configure an API key for at least one supported provider.
Use the CLI commands for taxonomy, mutation, judging, demos, or test runs.
Export results as JSON or HTML reports.

Verdict

PromptRAID is a useful fit for teams or researchers who need a self-hosted command-line framework for prompt-injection red teaming and reports of test outcomes. Look elsewhere if you need mutation generators for triggered prompt injection or data leakage, since those categories are not yet implemented.

Questions

Is PromptRAID free?

Yes. It is free and open source under the MIT License.

Which systems does it run on?

PromptRAID lists Linux and Windows. It requires Python 3.9 or later.

Which model providers can it use?

The harness supports Anthropic, Gemini, Groq, Cerebras, and OpenRouter, including OpenAI-compatible endpoints.

Does PromptRAID test data leakage?

Data leakage appears in its taxonomy, but the README says it does not yet have a mutation generator.

Who made PromptRAID?

The repository names Shikhali Jamalzade (alisalive), an independent offensive security researcher, as its author.

What does the MIT License allow?

It permits use, modification, distribution, and sale subject to its conditions. It provides the software as is and disclaims warranties and liability.

Facts from PromptRAID's own pages

ATLAS coverage: Implemented mutation categories map to verified MITRE ATLAS technique IDs, while some listed categories remain unverified placeholders or have no generator yet. · github.com, 9 Oct 2026
Installation: The README recommends installing the CLI with pipx and also documents installation in a Python virtual environment. · github.com, 10 Oct 2026
Intended users: The project describes its intended use as systematic red-team testing and benchmarking of LLM-based agents across providers. · github.com, 10 Oct 2026
Interface: The project provides a command-line interface with commands for taxonomy, mutation, judging, running tests, demos, and reports. · github.com, 9 Oct 2026
License: PromptRAID is distributed under the MIT License, which permits use, copying, modification, distribution, sublicensing, and sale subject to its terms. · github.com, 10 Oct 2026
Limitations: The README says triggered prompt injection and data leakage appear in its taxonomy but do not yet have mutation generators. · github.com, 10 Oct 2026
Maintainer: The README identifies Shikhali Jamalzade (alisalive) as an independent offensive security researcher and the project author. · github.com, 10 Oct 2026
Maker: The repository names Shikhali Jamalzade (alisalive) as the author. · github.com, 9 Oct 2026
Model providers: The harness supports Anthropic, Gemini, Groq, Cerebras, and OpenRouter, including OpenAI-compatible endpoints. · github.com, 9 Oct 2026
Payloads: Its mutation engine generates prompt injection variants using encoding tricks, role-play framing, delimiter breaking, and indirect injection through tool or document output. · github.com, 10 Oct 2026
Payload testing: Its mutation engine generates prompt injection variants using encoding tricks, jailbreak framing, delimiter breaking, and indirect injection through tool or document output. · github.com, 9 Oct 2026
Provider credentials: Running tests requires at least one provider API key, with environment variables documented for Anthropic, Gemini, Groq, OpenRouter, and Cerebras. · github.com, 10 Oct 2026
Providers: The harness supports Anthropic, Gemini, Groq, Cerebras, and OpenRouter, including OpenAI-compatible endpoints. · github.com, 10 Oct 2026
Purpose: PromptRAID is an automated prompt injection red-teaming framework for LLM-based agents, mapped to MITRE ATLAS. · github.com, 10 Oct 2026
Reports and storage: PromptRAID exports JSON and HTML reports and stores runs and per-payload results in SQLite. · github.com, 10 Oct 2026
Requirements: The package requires Python 3.9 or later and asks users to configure at least one supported provider API key. · github.com, 9 Oct 2026
Security classification: The project maps prompt injection tests to MITRE ATLAS and describes them as red-team tests for LLM agents. · github.com, 10 Oct 2026
Taxonomy: Mutation categories map to MITRE ATLAS technique IDs, which the project says it checks against the live ATLAS matrix. · github.com, 10 Oct 2026
Unverified categories: Three additional categories are described as unverified placeholders and are excluded from the listed ATLAS mapping. · github.com, 10 Oct 2026
Verdicts: The judge classifies results as success, partial compliance, fail, or inconclusive, and distinguishes neutral quoting from adopting an injected instruction. · github.com, 10 Oct 2026
Verdict scoring: The judge classifies results as success, partial compliance, fail, or inconclusive and distinguishes neutral echo from adopted compliance. · github.com, 9 Oct 2026
Warranty: The MIT License provides the software “as is” and disclaims warranties and liability. · github.com, 9 Oct 2026

Where PromptRAID stands on Freedom251

AI Security Testing Tools#21 of 29