Tagged prompt-injection

Red team LLMs with jailbreak prompt techniques and compare their guardrails

Extract structured data from web pages with PII and prompt injection checks

Scan untrusted input for prompt injection and unsafe content before a model sees it

Test prompt injection, retrieval poisoning and tool-boundary abuse in CTF agents

Test LLM apps and agents for prompt injection, tool abuse and RAG exposure

Flag prompt injection in LLM inputs using regex, heuristics and a DeBERTa classifier

Scan web pages, PDFs and OCR text for hidden prompt injection before agents read them

Probe LLM endpoints for jailbreaks, prompt injection and data leakage with garak

Test RAG pipelines for poisoned documents and indirect prompt injection

Assess AI and LLM systems for prompt injection, jailbreaks and tool abuse

Scan agent skills for malicious patterns, prompt injection and over-permissions

Add jailbreak, PII and toxicity guardrails to LLM apps with NeMo Guardrails and Colang

Detect prompt injection and jailbreak attempts in prompts and third-party data for RAG

Test LLM apps and agents for prompt injection, tool abuse, RAG poisoning and MCP risks

Test an AI agent for direct prompt injection and system role override attempts

Test LLM apps for direct and indirect prompt injection and prompt leaks

Hunt prompt injection and tool-call hijacking in AI-integrated applications

Hunt prompt injection and data exfiltration bugs in LLM-powered features

Scan agent skills, plugins and MCP servers for prompt injection and secret leaks

Assess AI and MCP systems for prompt injection, tool abuse and data leaks

Block prompt injection, PII leaks and social engineering in a mortgage assistant

Solve CTF challenges that involve attacking ML models, prompt injection and LoRA

Test LLM apps for prompt injection, indirect RAG attacks and MCP tool abuse

Solve CTF challenges on adversarial examples, model extraction and LLM jailbreaks

Test LLM apps for indirect prompt injection through web pages and documents

Test chatbots against jailbreak techniques such as role-play and encoding tricks

Test LLM apps for prompt injection, system prompt leaks and unsafe tool use

Harden LLM apps against prompt injection, tool-call abuse and unsafe outputs

Rank retrieved passages, drop irrelevant ones and flag hidden prompt injection

Defend AI agents against prompt injection, tool abuse and data theft

Run red team exercises for jailbreaks, data leaks and agent tool abuse

Defend AI systems from prompt injection using input controls and tool permissions

Inspect OpenClaw HTTP traffic for egress leaks and prompt injection via a proxy

Scan agent skills for prompt injection, malicious scripts and secret exposure