AIRTA Red Team

AIRTA Red Team is a self-hosted tool for authorized runtime security assessments of LLMs and LLM-driven applications. It is best suited to red teams, pentesters, and AppSec or MLsec teams seeking structured attack runs and evidence. Its broad assessment workflow costs nothing, but its deprecated, unmaintained status is a serious drawback for ongoing use.

Overview

The workflow builds suites from security playbooks, discovers target components and authentication, runs attacks, assesses outcomes, and exports reports. Because it evaluates observable runtime behavior, it can help teams identify issues exposed through application interactions; it does not perform source-code static analysis. Use it only on systems for which you have explicit permission.

Deployment can be local in Docker, behind a firewall, in CI, or against external endpoints for authorized black-box assessments. Targets can be reached with a Playwright browser runner or direct HTTP API transports, including file uploads and tool calls. This gives teams options for browser-based and API targets, while requiring them to supply their own LLM API key for test generation and assessment.

Key features

Playbooks and attack strategies

Playbooks cover OWASP LLM, OWASP Agent, MITRE ATLAS, jailbreak, system prompt exfiltration, and custom scenarios. Strategies include zero-shot, few-shot, multi-shot, iterative, multimodal, jailbreak, and chain-of-thought testing. This range supports varied runtime probes without limiting teams to a single attack style.

Execution evidence and reporting

Run artifacts capture prompts, uploads, responses, and tool calls in attack_log.json. The generated pipeline_report.json provides category rollups and severity, giving reviewers material to examine after a run. Results can also be exported through AIRTA Systems’ bulk import API for review and regression tracking.

Pricing

AIRTA Red Team: free. The open-source plan has no license fee or subscription, making it suitable for teams that can operate a self-hosted assessment tool. The main cost consideration is that you must bring your own LLM API key; no trial or paid tier is needed to access the free plan.

Platforms

AIRTA Red Team supports API, Linux, macOS, web, and Windows. Its self-hosted deployment options make it a fit for teams that want to run assessments in their own environments rather than rely on a hosted testing workflow.

Who it's for

Red teams, whitehat testers, pentesters, and AppSec or MLsec teams can use it to probe authorized LLM applications and preserve run evidence. It is less suitable for teams that need static code analysis, or for organizations that depend on active maintenance and support.

Pros and cons

• Pros: Playbooks and strategies span multiple attack styles, with custom playbooks and test cases for tailored assessments.
• Pros: Browser and direct HTTP execution support a range of target interfaces, while captured run artifacts and severity rollups aid review.
• Pros: The open-source plan has no license fee or subscription.
• Cons: Users must supply an LLM API key for test generation and assessment.
• Cons: Runtime assessment does not replace source-code static analysis.
• Cons: The repository marks the project deprecated and no longer maintained, a significant concern for continued security work.

Alternatives

For a broader comparison, see AI Security Testing Tools.

• Project Moonshot is another free option with API, Linux, macOS, self-hosted, web, and Windows support; consider it if those platform options better suit your deployment.
• OpenSecureAI Scanner has a free plan and a Pro plan at $49 a month, billed $49/month, with 50,000 API requests/month and 120 requests/minute per key; choose it if those stated Pro request limits and hosted AI remediation or LLM Gateway completions match your needs.
• PromptGuard offers a free plan with 20,000 scans a month, one API key, one project, and 24-hour log retention, plus a Team plan at $19 a month; it may suit readers whose needs fit those free-plan caps.
• Promptfoo’s free Community plan includes 10k red-team probes/month, LLM evaluation features, model providers and integrations, and local or self-hosted operation; choose it if that stated probe allowance and evaluation scope fit your workflow.
• PyRIT is a free open-source framework that requires a Python environment and configured AI endpoints; consider it if that framework setup suits your team.
• Augustus is free, open-source software for Linux, macOS, self-hosted environments, and Windows; API usage may require provider credentials.
• Basilisk is free AGPL-3.0 software distributed as a CLI, desktop app, Docker image, or source, for authorized use only.
• F5 BIG-IP APM is a paid alternative with a free trial and perpetual-license availability across virtual editions, hardware, or hybrid environments.

Verdict

AIRTA Red Team is a sensible no-license-fee option for security teams that want a self-hosted runtime workflow with diverse attack strategies and inspectable run artifacts. Its key drawback is decisive for long-term adoption: the project is deprecated and no longer maintained. Choose it for authorized assessments only if your team can accept that maintenance risk and supply its own LLM API key; otherwise, consider an alternative better aligned with your support and workflow needs.