OverServe: Resource Amplification Attacks on SaaS Agents via Tool--Prompt Co-Design
Organizations: Florida State University · Johns Hopkins University
Abstract
Advances in large language models (LLMs) are enabling software-as-a-service (SaaS) providers to automate longer business workflows with agents. In subscription-based SaaS deployments, model and tool calls consume provider resources without additional customer payments. Because the agent autonomously decides when to stop, a malicious tenant can send a prompt to the agent, exploiting its completion decisions to extend execution and waste provider resources. In this paper, we present OverServe, a multi-agent framework for constructing such attacks against SaaS agents. OverServe co-designs Model Context Protocol (MCP) status tools and completion-confirmation prompts for the attacker. Each tool call returns accurate status and terminates normally, yet the agent may repeatedly check an unfinished condition to satisfy the prompt. This consumes the SaaS provider's resources even after the original business task is completed. OverServe constructs service-specific tool--prompt pairs and uses separate tool and prompt reviews to guide candidate pair selection. Tested tool specifications trigger no alerts in the evaluated MCP scanners. We evaluate Qwen, Nemotron, DeepSeek, and Sonnet on selected tasks from nine benchmarks. All OverServe runs pass the original task checks. OverServe can reach a maximal amplification of API requests by while preserving the original task goal and collateral checks.
Figures & tables
| Qwen3.8-27B | Nemotron-3-Super | DeepSeek-v4-Flash | Sonnet-5 | |||||||||||||
| Benchmark | Resource use | Amp. | S/P/D | Hit | Resource use | Amp. | S/P/D | Hit | Resource use | Amp. | S/P/D | Hit | Resource use | Amp. | S/P/D | Hit |
| MCPMark | ||||||||||||||||
| TxtMerge | .85/.90/1 | .95/.85/1 | .95/.90/1 | .98/.80/1 | ||||||||||||
| Recommender * | .90/.90/1 | .95/.85/1 | .95/.90/1 | .98/.80/1 | ||||||||||||
| EnterpriseOps-Gym | 40.00 | 1/.80/1 | .98/.80/1 | .95/.90/1 | .98/.90/1 | |||||||||||
| BrowseComp-Plus | Baseline task not completed | .98/.85/1 | ||||||||||||||
| MCPMark: TxtMerge | EnterpriseOps-Gym | Stealth Review | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Qwen3.8-27B | Sonnet-5 | Qwen3.8-27B | Sonnet-5 | ||||||||
| Attack | Quiet | Churn | Quiet | Churn | Quiet | Churn | Quiet | Churn | Code | Runtime | Prompt |
| A2M (return port) | 0/2 | 0/2 | original | ||||||||
| LoopTrap | 0/2 | 0/2 | original | ||||||||
| STA (4 rounds) | NA | NA | 4/4 | ||||||||
| OverServe | 2/2 | 2/2 | 2/2 | ||||||||
| (A) Prompt controls: fixed tool | ||||
|---|---|---|---|---|
| Quiet | Churn | |||
| Prompt | All | Crafted | All | Crafted |
| Original task prompt | 4 | 0 | 4 | 0 |
| Unrelated prompt | 27 | 0 | 25 | 0 |
| Matched prompt | 5 | 1 | 40 | 36 |
| Paraphrased prompt | 5 | 1 | 40 | 36 |
| Industrial scanner | Research detectors | ||
|---|---|---|---|
| Scanners | Alerts | Detector | Alerts |
| Cisco YARA ( Cisco AI Defense, 2026 ) | 0/22 | DCI | 3/22 |
| Lasso regex ( Lasso Security, 2025 ) | 0/22 | Connor | 1/22 |
| MCPInspect ( Li and Gao, 2026 ) | 0/22 | Repeat-cue | 0/22 |
| MCPShield ( Nikita, 2025 ) | 0/22 | No fixed point | 6/22 |
| LLM tool review | 0/22 | ||
Appendix figures & tables5 assets
Supplementary material from the paper’s appendix.
Appendix
| Setting | Task subset | Execution limit | Baseline / attacked | Verification |
|---|---|---|---|---|
| MCPMark | TxtMerge; earlier replay | 100 model turns; earlier 50 | Updated TxtMerge: 2/3; earlier Q: 3/3, N,D,S: 1/1 | Native task checks |
| EOG | Channel and message | 40 or 200 iterations | Selected Q: 2/2; others: 1/1 | Channel and associated message |
| BrowseComp-Plus | Query 655 | Adapter limit | 1/1 per G/L pair | Graded answer |
| STATE-Bench | Three support tasks | 30 status reads; override | Frozen Q: 6/6; S: 5/5 valid. Search Q: 6/6; N,D,S: 3/3 | Required-state assertions |
| TheAgentCompany | Install Go | 40 iterations | Q,N: 2/2; D,S: 2/3 | Two native checkpoints |
| Toolathlon | Support review | 40 model turns | Q: 1/1; N,D,S: 1/2 | Native completion/evaluation |
| Quiet | Churn | |||
|---|---|---|---|---|
| Method | Total | Reuse | Total | Reuse |
| OverServe | 1.21 | 1.02 | 8.70 | 1.62 |
| A2M (return port) | 1.21 | 1.03 | 1.21 | 1.03 |
| LoopTrap-style | 2.92 | 1.30 | 4.97 | 1.30 |
| STA (4 rounds) | 1.32 | 1.04 | 1.23 | 1.04 |
| Victim | Clean | Quiet | Churn | Amp. | Pass |
|---|---|---|---|---|---|
| Sonnet-5 | 4,4 | 5,5 | 9,12 | 2.10 | 2/2 |
| DeepSeek v4.1 | 5,10 | 6,6 | 14,52 | 5.50 | 2/2 |
| Gemini | 13,13 | 13,10 | 25,29 | 2.41 | 2/2 |
| Qwen | 5,5 | 8,6 | 51,51 | 7.44 | 6/6 |
| Deployment | Pairs | Quiet | Churn | Amp. | Pass | Inv. | Limit |
|---|---|---|---|---|---|---|---|
| Qwen3.8-27B | 6/6 | 5.5 | 19.5 | 6/6 | 3/6 | 3/6 | |
| Sonnet-5 | 5/6 | 5.0 | 11.0 | 2/5 | 4/5 | 0/5 | |
| Gemini-3.8-Flash | 12/12 | 5.0 | 8.0 | 11/12 | 6/12 | 0/12 | |
| DeepSeek v4.1-Flash | 6/6 | 5.5 | 9.0 | 3/6 | 3/6 | 0/6 |
LLM Agent Security
Large language model agent security studies attacks and safeguards for systems that plan, use tools, retain memory, and act across applications.
See all 351 papers