Deconvolute Labs
deconvolute-labs
The Application Firewall for the Model Context Protocol (MCP). Protects Agents and RAG pipelines from Confused Deputy attacks and indirect injection.
Languages
Top Repositories
Real-time security layer protecting AI Agents from Confused Deputy attacks, malicious MCP payloads, and Indirect Prompt Injection.
A demonstration of RAG poisoning attacks using dormant documentation injections.
Reproducible security benchmarking for the Deconvolute SDK and AI system integrity against adversarial attacks.
Live PoC: MCP rug pull attack that steals AI agent credentials mid-session and how to block it in 3 lines of code.
Automatically generate YARA rules from adversarial and benign text samples. Built for detecting indirect prompt injection attacks on RAG pipelines.
Repositories
6Real-time security layer protecting AI Agents from Confused Deputy attacks, malicious MCP payloads, and Indirect Prompt Injection.
Live PoC: MCP rug pull attack that steals AI agent credentials mid-session and how to block it in 3 lines of code.
A demonstration of RAG poisoning attacks using dormant documentation injections.
Automatically generate YARA rules from adversarial and benign text samples. Built for detecting indirect prompt injection attacks on RAG pipelines.
Reproducible security benchmarking for the Deconvolute SDK and AI system integrity against adversarial attacks.
No description provided.