Skip to content
SPECIFICTION
Get listed

RedForge

github.com

Free plan LinuxAPISelf-hosted
Visit github.com
Spec score 7.7 #17 of 29 in AI Security Testing Tools

What it is

RedForge is a free, open-source LLM security testing platform for red teaming and vulnerability scanning. It runs on Linux, through an API and as self-hosted software. The project lists more than 49 probes, full OWASP LLM Top 10 coverage and YAML probe additions without Python. Attack modules include PAIR, Crescendo multi-turn escalation, Skeleton Key, many-shot jailbreaking and GCG adversarial suffixes. Its mutation engine expands two payloads into 37 variants using encoding, Unicode, cognitive reframing and structural injection strategies. Provider adapters cover OpenAI, Anthropic, Google Gemini, AWS Bedrock, Azure OpenAI, Mistral, Ollama, HuggingFace and generic REST endpoints. Ollama runs locally without an API key. Scans create JSON, JSONL audit and failures reports, with HTML, Markdown and SARIF also available. Compliance reports cover NIST AI RMF, EU AI Act, ISO/IEC 42001, GDPR and OWASP LLM Top 10. RedForge is Apache 2.0 software for authorized use.

Spec sheet

Purpose
RedForge is an LLM security testing platform for red teaming and vulnerability scanning.Read 9 Oct 2026 · github.com
Deployment
The Docker instructions say the container runs as non-root user redforge (UID 1001) with a read-only filesystem.Read 9 Oct 2026 · github.com
Integrations
It can export SARIF for GitHub Code Scanning and documents a GitHub Actions workflow that uploads the SARIF file.Read 9 Oct 2026 · github.com
Compliance
Compliance reporting covers NIST AI RMF, EU AI Act, ISO/IEC 42001, GDPR, and OWASP LLM Top 10, and users can add custom YAML frameworks.Read 9 Oct 2026 · github.com
License
The project is released under the Apache 2.0 license, and the README says probe payloads are released under CC0.Read 9 Oct 2026 · github.com
Attack methods
Attack modules include PAIR, Crescendo multi-turn escalation, Skeleton Key, many-shot jailbreaking, and GCG adversarial suffixes.Read 9 Oct 2026 · github.com
Attack modules
Its attack modules include PAIR, Crescendo multi-turn escalation, Skeleton Key, many-shot jailbreaking, and GCG adversarial suffixes.Read 9 Oct 2026 · github.com
CI integration
RedForge can emit SARIF for GitHub Code Scanning and exits with status 1 on critical or high findings for use as a CI gate.Read 9 Oct 2026 · github.com
Detectors
The README lists seven detector types, including YARA-style, similarity, refusal, regex, keyword, code, and unsafe-content detectors.Read 9 Oct 2026 · github.com
Guardrails
A bidirectional guardrail pipeline scans inputs and outputs for prompt injection, PII, toxicity, and secrets.Read 9 Oct 2026 · github.com
License and use
The repository identifies the license as Apache 2.0 and says the tool is for authorized use only.Read 9 Oct 2026 · github.com
Local use
The Ollama adapter runs locally without an API key, and the README describes it as zero cost.Read 9 Oct 2026 · github.com
Model providers
It lists adapters for OpenAI, Anthropic, Google Gemini, AWS Bedrock, Azure OpenAI, Mistral, Ollama, HuggingFace, and generic REST endpoints.Read 9 Oct 2026 · github.com
Mutation engine
The mutation engine expands two payloads to 37 variants across encoding, Unicode, cognitive reframing, and structural injection strategies.Read 9 Oct 2026 · github.com
Probe coverage
It lists 49+ probes and full OWASP LLM Top 10 coverage, with YAML probes that can be added without Python.Read 9 Oct 2026 · github.com
Providers
The README lists provider adapters for OpenAI, Anthropic, Google Gemini, AWS Bedrock, Azure OpenAI, Mistral, Ollama, HuggingFace, and generic REST endpoints.Read 9 Oct 2026 · github.com
Reports
Scans automatically produce JSON, JSONL audit, and failures reports, with additional HTML, Markdown, SARIF, JSON, audit, and failures formats.Read 9 Oct 2026 · github.com
Security controls
Documented controls include masking API keys in logs, setting result files to mode 0600, bearer-token REST API authentication, disabled-by-default CORS, and a rate limit of 10 scan requests per minute per client.Read 9 Oct 2026 · github.com
Usage requirement
The CLI requires an authorization value indicating owned, authorized, or research use.Read 9 Oct 2026 · github.com

Plans and prices

RedForge Free Apache 2.0 open-source software · install with pip Read 9 Oct 2026 · github.com

Every RedForge plan and price →

Strong points and limits

Strong points

  • More than 49 probes with OWASP LLM Top 10 coverage
  • Supports many hosted and local model providers
  • Generates SARIF for GitHub Code Scanning
  • Custom YAML probes and compliance frameworks
  • Ollama runs locally without an API key

Limits

  • Authorized-use value is required by the CLI
  • Linux, API and self-hosted platforms only
  • Rate limit is 10 scan requests per minute per client

Who it suits

RedForge suits security teams and developers performing authorized LLM red teaming, vulnerability scans or CI checks. It fits users who want self-hosted testing and configurable probes.

Verdict

RedForge provides broad, configurable LLM security testing at no listed charge. Its self-hosted design, provider adapters and SARIF output support both local work and CI workflows.

Where Specifiction places it

Alternatives to RedForgeAll alternatives →