OverServe: Resource Amplification Attacks on SaaS Agents via Tool--Prompt Co-Design

Summary

OverServe is a multi-agent framework that builds attacks on SaaS agents by co-designing MCP status tools and completion-confirmation prompts. The tool calls return accurate status and terminate normally, yet the agent repeatedly checks an unfinished condition, consuming provider resources after the original task is done. Tested tools triggered no alerts in the evaluated MCP scanners, and OverServe reached a maximal amplification of API requests by 93.88x across Qwen, Nemotron, DeepSeek, and Sonnet.

Topics

• AI agents
• Model Context Protocol

Related items

• MediumEndor Labs’ AI SAST Finds Zero Day Memory-Amplification DoS in Anthropic’s buffa library | Endor LabsSimilar attack · Endor Labs Blog
• MediumCVE-2026-44627: Intel oneDNN integer overflow to buffer overflow in Ring 3 user applicationsSimilar attack · NVD/CVE Database
• MediumGHSA-h7c4-q426-7jh9: tsParticles: A single anonymous MCP session can monopolize HTTP mode and deny new clientsSimilar attack · GitHub Advisory Database
• HighGHSA-4p8p-pjpq-2w7h: tsParticles: Malformed JSON requests can terminate the @tsparticles/mcp-server HTTP processSimilar attack · GitHub Advisory Database
• HighGHSA-447x-qqwf-m7xq: tsParticles: Unbounded HTTP Request Body Buffering in @tsparticles/mcp-server Leads to Remote DoSSimilar attack · GitHub Advisory Database