LLMMap review
Focused prompt-injection testing with broad attack coverage and flexible LLM backends.
Reviewed by iTechGuides Editors · Editorial team · Updated Oct 2026
LLMMap is an open-source, self-hosted command-line framework for authorized security testing of applications that incorporate language models. It is aimed at web security teams assessing prompt-injection vulnerabilities across HTTP requests. The framework discovers injection points in query parameters, request bodies, headers, cookies, and paths, then sends targeted probes and analyzes the resulting responses. Direct URL targets and Burp Suite XML request imports support common testing workflows.
Its strongest area is prompt-injection test depth. LLMMap includes 227 techniques across 18 attack families and generates goal-directed attack prompts through a dual-LLM generator-and-judge workflow. Obfuscation options include base64, homoglyph, leetspeak, and language switching. Safe mode and dry-run scanning provide controls for planning and limiting probe activity. The framework also supports prompt-injection, jailbreak, and data-leakage tests, giving teams several related test objectives within its focused scope.
LLMMap fits teams that want an open-source framework they can run locally and connect to different model backends. Supported options include Ollama, OpenAI, Anthropic, and Google, while Burp Suite integration helps connect application testing with request capture and replay workflows. Findings are confirmed through repeated probes and Wilson confidence intervals rather than a single response. Teams seeking focused prompt-injection assessment should find the scope well aligned; teams needing broader application security coverage or a hosted product should look for a different tool. LLMMap is best suited to security practitioners comfortable with self-hosted command-line workflows and authorized testing environments.
LLMMap pros and cons
- Where it wins
- 227 prompt-injection techniques across 18 attack families
- Burp Suite imports plus direct URL targets
- Repeated probes with Wilson confidence intervals
- Where it doesn't
- Focused on prompt injection rather than broad security testing
- Self-hosted command-line deployment
- Intended for authorized testing of LLM applications
LLMMap fact sheet, pricing and score →
Advertiser disclosure: iTechGuides is reader-supported. Vendors can pay for top positions in our rankings and for a place on other products' pages, and we may earn a commission when you click some links. Bought places stand at the number bought; this page and our disclosure say so. Paying vendors may get a fuller, more detailed review from our editors, and that review can change their score. How we rank.
Last updated · How we research and update