Paper Citation Record · LEDGER

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems

As of 17 August 2026, this Paper Citation Record lists 100 of 107 outbound references and 10 inbound Pith citation observations for arXiv:2506.13666.

A citation records a reference. It does not transfer a finding from one paper to another.

pith.paper-citation-record.v1
2506.13666 v1

Coverage vector

measured 100 of 107 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-07T00:32:36.760608Z

measured 110 of 110 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-17T06:30:58.91139+00:00

measured 10 of 10 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-06T21:44:54.864170Z

measured 1 of 1 external citation measurements

A source-named dated measurement, never combined with another source.

Source: arxiv_reference, observed 2026-08-05T02:28:24.338817Z

Reference resolution

100 of 107 outbound references displayed

verified exact0
verified fuzzy13
unresolved87
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

0
arxiv_reference, observed 2026-08-05T02:28:24.338817Z

Outbound references

Observation cab8fd2c-7197-4197-b0fc-fef0587b23a5 · outbound

This paper cites Is My Data in Your Retrieval Database? Membership Inference Attacks Against Retrieval Augmented Generation.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Is My Data in Your Retrieval Database? Membership Inference Attacks Against Retrieval Augmented Generation

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.443050Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation da9cb528-7157-4b44-8b73-00c0227eb123 · outbound

This paper cites Introducing the model context protocol.https://www.anthropic.com/news/ model-context-protocol, November 2024.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Introducing the model context protocol.https://www.anthropic.com/news/ model-context-protocol, November 2024

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.447486Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e62633f9-e83d-4a72-b74f-41265a3c4fd3 · outbound

This paper cites Foundational Challenges in Assuring Alignment and Safety of Large Language Models.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Foundational Challenges in Assuring Alignment and Safety of Large Language Models

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.451035Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation db8f1793-bf40-42a1-9c1f-9efc77144c16 · outbound

This paper cites Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.454938Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e8dd5d7a-7962-493e-b320-9fbace30b48a · outbound

This paper cites Safety-tuned LLaMAs: Lessons from improving the safety of large language models that follow instructions.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Safety-tuned LLaMAs: Lessons from improving the safety of large language models that follow instructions

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.458413Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1a387885-bd0a-4de2-9931-609e4c15616b · outbound

This paper cites an unresolved cited work.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Unresolved cited work

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.461485Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 89437d5c-6824-41ee-9761-7ae1f94654d8 · outbound

This paper cites Highlights from lex fridman’s interview of yann lecun, March.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Highlights from lex fridman’s interview of yann lecun, March

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.464997Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c256f18c-5b5b-4146-88f5-2b3a8045b973 · outbound

This paper cites A survey on evaluation of large language models.ACM transactions on intelligent systems and technology, 15(3):1–45, 2024.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems A survey on evaluation of large language models.ACM transactions on intelligent systems and technology, 15(3):1–45, 2024

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.471713Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 11cc4119-caec-474e-968a-138ed5374a77 · outbound

This paper cites Pappas, Florian Tramèr, Hamed Hassani, and Eric Wong.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Pappas, Florian Tramèr, Hamed Hassani, and Eric Wong

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.474958Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4c2d0d40-1d6a-4979-80f0-8940a1819874 · outbound

This paper cites Agentpoison: Red-teaming llm agents via poisoning memory or knowledge bases.Advances in Neural Information Processing Systems, 37:130185–130213, 2025.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Agentpoison: Red-teaming llm agents via poisoning memory or knowledge bases.Advances in Neural Information Processing Systems, 37:130185–130213, 2025

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.477590Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 599af7df-89b2-426a-9af3-6f744f976cd4 · outbound

This paper cites Safety-aware fine-tuning of large language models.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Safety-aware fine-tuning of large language models

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.480493Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 349a1758-8d22-4163-8e06-8da4d2fdd95d · outbound

This paper cites Scaling instruction-finetuned language models.Journal of Machine Learning Research, 25(70):1–53, 2024.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Scaling instruction-finetuned language models.Journal of Machine Learning Research, 25(70):1–53, 2024

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.483792Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4cb8a155-06a1-4f7e-ac1c-f3da080f6365 · outbound

This paper cites Textworld: A learning environment for text-based games.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Textworld: A learning environment for text-based games

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.486870Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 47c933f4-d52b-4cd3-a619-3aae05a6b0ad · outbound

This paper cites DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.489432Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b12f1fb7-e970-4920-862b-1ac01661ce5c · outbound

This paper cites Ai agents under threat: A survey of key security challenges and future pathways.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Ai agents under threat: A survey of key security challenges and future pathways

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.492579Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5d55cfb2-91e3-4329-b16a-8462d561c64c · outbound

This paper cites Bert: Pre-training of deep bidirectional transformers for language understanding.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Bert: Pre-training of deep bidirectional transformers for language understanding

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.495577Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e7069162-a4b2-4c9d-8d8a-ab81d166a0c9 · outbound

This paper cites A Wolf in Sheep's Clothing: Generalized Nested Jailbreak Prompts can Fool Large Language Models Easily.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems A Wolf in Sheep's Clothing: Generalized Nested Jailbreak Prompts can Fool Large Language Models Easily

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.498449Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8888b506-9c6e-4569-bb1b-e1f604fda046 · outbound

This paper cites A survey of agent interoperability protocols: Model Context Protocol (MCP), Agent Communication Protocol (ACP), Agent-to-Agent Protocol (A2A), and Agent Network Protocol (ANP).

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems A survey of agent interoperability protocols: Model Context Protocol (MCP), Agent Communication Protocol (ACP), Agent-to-Agent Protocol (A2A), and Agent Network Protocol (ANP)

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.501355Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 82bac543-6e04-41e9-8fa1-0d01ddab0f31 · outbound

This paper cites Pawan Kumar, and Adel Bibi.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Pawan Kumar, and Adel Bibi

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.504729Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 87ea844c-4b84-4efe-b22f-b0f8cc12c09b · outbound

This paper cites Imprompter: Tricking LLM Agents into Improper Tool Use.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Imprompter: Tricking LLM Agents into Improper Tool Use

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.507283Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b2baea63-0248-4742-94a6-3ba85ab31972 · outbound

This paper cites Red Teaming Language Models to Reduce Harms: Methods, Scaling Behaviors, and Lessons Learned.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Red Teaming Language Models to Reduce Harms: Methods, Scaling Behaviors, and Lessons Learned

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.510129Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 71d669d1-12cd-4a41-869c-ffc4f118f42d · outbound

This paper cites Textbooks Are All You Need.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Textbooks Are All You Need

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.512749Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6338beed-56ee-4479-b1ce-ca417488e76c · outbound

This paper cites Large Language Model based Multi-Agents: A Survey of Progress and Challenges.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Large Language Model based Multi-Agents: A Survey of Progress and Challenges

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.515911Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 15daf63a-56cd-4b96-afdb-15cb2ff8a771 · outbound

This paper cites Regulating chatgpt and other large generative ai models.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Regulating chatgpt and other large generative ai models

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.519516Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 74fbc7fb-4214-4ff9-92f1-0361627e54d6 · outbound

This paper cites A survey on large language models: Applications, challenges, limitations, and practical usage.Authorea Preprints, 2023.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems A survey on large language models: Applications, challenges, limitations, and practical usage.Authorea Preprints, 2023

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.522302Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c9306aa5-16ba-4a9d-985a-e15869da3867 · outbound

This paper cites What is in Your Safe Data? Identifying Benign Data that Breaks Safety.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems What is in Your Safe Data? Identifying Benign Data that Breaks Safety

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.525213Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 73b9fd81-05b3-489c-b34b-cac9663ac932 · outbound

This paper cites Red-Teaming LLM Multi-Agent Systems via Communication Attacks.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Red-Teaming LLM Multi-Agent Systems via Communication Attacks

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.528134Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ed2a81ee-4b17-4cac-8223-bbed55d65379 · outbound

This paper cites Model Context Protocol (MCP): Landscape, Security Threats, and Future Research Directions.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Model Context Protocol (MCP): Landscape, Security Threats, and Future Research Directions

Reference 29

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.535179Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ecd65069-43f6-4e93-b30d-fac8feff78d3 · outbound

This paper cites T1: Advancing Language Model Reasoning through Reinforcement Learning and Inference Scaling.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems T1: Advancing Language Model Reasoning through Reinforcement Learning and Inference Scaling

Reference 30

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.538291Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3ed7c96d-4758-44d2-816c-28d632e2b569 · outbound

This paper cites Scaling Trends in Language Model Robustness.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Scaling Trends in Language Model Robustness

Reference 31

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.541321Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 740059a3-efc9-4b55-8bda-5f99777f08ed · outbound

This paper cites A survey of safety and trustworthiness of large language models through the lens of verification and validation.Artificial Intelligence Review, 57(7):175, 2024.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems A survey of safety and trustworthiness of large language models through the lens of verification and validation.Artificial Intelligence Review, 57(7):175, 2024

Reference 32

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.544502Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation abacfcb5-873b-48ac-93bf-9f014ab7eb00 · outbound

This paper cites Babyai 1.1, 2020.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Babyai 1.1, 2020

Reference 33

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.547224Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 427cf786-071f-45a2-b516-8bde402013dc · outbound

This paper cites Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversations.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversations

Reference 34

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.550081Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ec439bd0-f2ff-4c1a-a184-71b64c84c100 · outbound

This paper cites Mcp security notification: Tool poisoning attacks.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Mcp security notification: Tool poisoning attacks

Reference 35

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.553330Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1244253f-4154-4114-bd33-9b4ed1d2517d · outbound

This paper cites SafeChain: Safety of Language Models with Long Chain-of-Thought Reasoning Capabilities.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems SafeChain: Safety of Language Models with Long Chain-of-Thought Reasoning Capabilities

Reference 36

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.556165Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bdb8422b-033e-43a6-89cf-c0bda4e306b9 · outbound

This paper cites Flooding Spread of Manipulated Knowledge in LLM-Based Multi-Agent Communities.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Flooding Spread of Manipulated Knowledge in LLM-Based Multi-Agent Communities

Reference 37

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.559678Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 19c5409f-c863-4e40-9a1b-ee970cc44425 · outbound

This paper cites Llm-mod: Can large language models assist content moderation? InExtended Abstracts of the CHI Conference on Human Factors in Computing Systems.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Llm-mod: Can large language models assist content moderation? InExtended Abstracts of the CHI Conference on Human Factors in Computing Systems

Reference 38

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.563012Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7f1e61ea-52c3-40fa-a6f5-62f2bf479779 · outbound

This paper cites Watch your language: Investigating content moderation with large language models.Proceedings of the International AAAI Conference on Web and Social Media, 2024.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Watch your language: Investigating content moderation with large language models.Proceedings of the International AAAI Conference on Web and Social Media, 2024

Reference 39

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.565644Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 511c0034-a37f-4235-b47d-765d0119c364 · outbound

This paper cites MCP Guardian: A Security-First Layer for Safeguarding MCP-Based AI System.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems MCP Guardian: A Security-First Layer for Safeguarding MCP-Based AI System

Reference 40

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.568714Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1d46beab-1ba3-4821-9853-c92ee7fc1333 · outbound

This paper cites How not to be stupid about ai, with yann lecun.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems How not to be stupid about ai, with yann lecun

Reference 41

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.571929Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4a01d0f6-66ab-4bb4-bdc9-77680eb4f755 · outbound

This paper cites Commercial LLM Agents Are Already Vulnerable to Simple Yet Dangerous Attacks.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Commercial LLM Agents Are Already Vulnerable to Simple Yet Dangerous Attacks

Reference 42

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.575227Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 59802d7a-0a03-4b0c-9a71-8ec832f81fca · outbound

This paper cites Common 7B Language Models Already Possess Strong Math Capabilities.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Common 7B Language Models Already Possess Strong Math Capabilities

Reference 43

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.578494Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c8132a73-f252-428c-96ae-e33004d5b5c2 · outbound

This paper cites Camel: Communicative agents for" mind" exploration of large language model society.Advances in Neural Information Processing Systems, 36:51991–52008, 2023.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Camel: Communicative agents for" mind" exploration of large language model society.Advances in Neural Information Processing Systems, 36:51991–52008, 2023

Reference 44

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.581487Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9a00be52-d04d-4e03-ab77-9f90bda7a673 · outbound

This paper cites DeepInception: Hypnotize Large Language Model to Be Jailbreaker.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems DeepInception: Hypnotize Large Language Model to Be Jailbreaker

Reference 45

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.584165Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a299b3ba-ca64-428c-a5f5-87a036e037af · outbound

This paper cites The Unlocking Spell on Base LLMs: Rethinking Alignment via In-Context Learning.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems The Unlocking Spell on Base LLMs: Rethinking Alignment via In-Context Learning

Reference 46

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.587266Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation feffbfcf-e616-4ee4-a8e2-f837a306ebf3 · outbound

This paper cites Understanding and enhancing the transferability of jailbreaking attacks.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Understanding and enhancing the transferability of jailbreaking attacks

Reference 47

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.590175Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5f27bde5-03c1-4068-bc86-5710b3f6facc · outbound

This paper cites ToolACE: Winning the Points of LLM Function Calling.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems ToolACE: Winning the Points of LLM Function Calling

Reference 48

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.593020Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 64067349-a619-4bd7-b98d-afd287c03953 · outbound

This paper cites Autodan: Generating stealthy jailbreak prompts on aligned large language models.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Autodan: Generating stealthy jailbreak prompts on aligned large language models

Reference 49

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.595996Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f578c017-50af-4261-9762-8f77c9152e97 · outbound

This paper cites Jailbreaking ChatGPT via Prompt Engineering: An Empirical Study.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Jailbreaking ChatGPT via Prompt Engineering: An Empirical Study

Reference 50

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.598721Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bbfc070e-c4aa-4838-bdd6-01e020769a5f · outbound

This paper cites RobustFT: Robust Supervised Fine-tuning for Large Language Models under Noisy Response.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems RobustFT: Robust Supervised Fine-tuning for Large Language Models under Noisy Response

Reference 51

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.601660Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f700f20c-b668-4c99-a7fc-334e65e65551 · outbound

This paper cites CodeChameleon: Personalized Encryption Framework for Jailbreaking Large Language Models.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems CodeChameleon: Personalized Encryption Framework for Jailbreaking Large Language Models

Reference 52

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.604717Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7954f0b0-91d6-43a7-8bd6-2edf9f2652f8 · outbound

This paper cites Agentboard: An analytical evaluation board of multi-turn llm agents, 2024.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Agentboard: An analytical evaluation board of multi-turn llm agents, 2024

Reference 53

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.607632Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f5dc2dfc-0fac-4b1a-b621-b202218ea1a0 · outbound

This paper cites Safety at Scale: A Comprehensive Survey of Large Model and Agent Safety.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Safety at Scale: A Comprehensive Survey of Large Model and Agent Safety

Reference 54

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.610249Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7b9f233f-ddce-4c8d-9b52-39901c78a083 · outbound

This paper cites AgentSafe: Safeguarding Large Language Model-based Multi-agent Systems via Hierarchical Data Management.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems AgentSafe: Safeguarding Large Language Model-based Multi-agent Systems via Hierarchical Data Management

Reference 55

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.613450Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c7de98e6-ed07-46f7-a216-adaf5a45e658 · outbound

This paper cites an unresolved cited work.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Unresolved cited work

Reference 56

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.617161Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4540d3ec-8f76-41be-b476-e4d565fa9ffe · outbound

This paper cites A Comprehensive Overview of Large Language Models.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems A Comprehensive Overview of Large Language Models

Reference 57

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.619901Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 26ca70b2-80f0-4e90-9655-1d19342371e8 · outbound

This paper cites GPT-4 technical report.CoRR, 2023.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems GPT-4 technical report.CoRR, 2023

Reference 58

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.623242Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 93a25dfa-c201-4bb1-aaef-b423ed7d83e6 · outbound

This paper cites Training language models to follow instructions with human feedback.Advances in neural information processing systems, 35:27730–27744, 2022.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Training language models to follow instructions with human feedback.Advances in neural information processing systems, 35:27730–27744, 2022

Reference 59

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.626365Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5153d8fc-048f-4b5a-a9a7-4b814de674f6 · outbound

This paper cites Self-alignment of large language models via monopolylogue-based social scene sim- ulation.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Self-alignment of large language models via monopolylogue-based social scene sim- ulation

Reference 60

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:32:37.618797Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 9e7ec986-2242-46b0-b1ce-329bf2db8fb7 · outbound

This paper cites A survey on agent-based modelling assisted by machine learning.Expert Systems, 42(1):e13325, 2025.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems A survey on agent-based modelling assisted by machine learning.Expert Systems, 42(1):e13325, 2025

Reference 61

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:32:37.608506Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation b9f45d87-fa24-49cb-852f-836ec4551680 · outbound

This paper cites ADaPT: As-Needed Decomposition and Planning with Language Models.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems ADaPT: As-Needed Decomposition and Planning with Language Models

Reference 62

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.635903Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bfe99cd4-0424-4e95-bcd6-a7d4a6e6dd80 · outbound

This paper cites Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!

Reference 63

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.639515Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fc79b600-34e2-42f3-bf03-69748daac3d7 · outbound

This paper cites Safety alignment should be made more than just a few tokens deep.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Safety alignment should be made more than just a few tokens deep

Reference 64

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:32:37.598823Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 54ba21df-c8af-47e0-a521-de0985373cb3 · outbound

This paper cites MCP Safety Audit: LLMs with the Model Context Protocol Allow Major Security Exploits.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems MCP Safety Audit: LLMs with the Model Context Protocol Allow Major Security Exploits

Reference 65

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.645879Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a26c16fc-22fc-46ef-83f1-d36f03a32ea9 · outbound

This paper cites Tptu: Task planning and tool usage of large language model-based ai agents.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Tptu: Task planning and tool usage of large language model-based ai agents

Reference 66

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:32:37.588649Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation eee225d5-1ecd-48db-83f4-59c3f3eb83e2 · outbound

This paper cites Toolformer: Language models can teach themselves to use tools.Advances in Neural Information Processing Systems, 36: 68539–68551, 2023.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Toolformer: Language models can teach themselves to use tools.Advances in Neural Information Processing Systems, 36: 68539–68551, 2023

Reference 67

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.651630Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 95941cb5-c39b-4245-9ab4-f38416c441f1 · outbound

This paper cites Large Language Model Safety: A Holistic Survey.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Large Language Model Safety: A Holistic Survey

Reference 69

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.657747Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e38e1f26-8af7-43dc-846e-f22688ba3362 · outbound

This paper cites Welcome to the era of experience.Google AI, 2025.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Welcome to the era of experience.Google AI, 2025

Reference 70

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:32:37.572748Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation bc18e7db-d374-4fb0-bf7f-cce0c18a9f5c · outbound

This paper cites Large language model (chatgpt) as a support tool for breast tumor board.NPJ Breast Cancer, 9(1):44, 2023.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Large language model (chatgpt) as a support tool for breast tumor board.NPJ Breast Cancer, 9(1):44, 2023

Reference 71

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:32:37.563133Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 068ee5fd-9a41-438b-bcaa-82dc516b2131 · outbound

This paper cites ToolAlpaca: Generalized Tool Learning for Language Models with 3000 Simulated Cases.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems ToolAlpaca: Generalized Tool Learning for Language Models with 3000 Simulated Cases

Reference 72

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.666302Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f034a09e-e2fa-4274-ab49-808463f0e4b1 · outbound

This paper cites A Survey on Post-training of Large Language Models.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems A Survey on Post-training of Large Language Models

Reference 73

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.669532Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e49dc2bf-5595-4833-a4fd-4f0d3ee92e48 · outbound

This paper cites Multi-Agent Collaboration Mechanisms: A Survey of LLMs.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Multi-Agent Collaboration Mechanisms: A Survey of LLMs

Reference 74

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.672515Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 036f6ddd-16e2-4554-a054-a056ef1b0c3c · outbound

This paper cites House Committee on Oversight and Accountability.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems House Committee on Oversight and Accountability

Reference 75

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:32:37.553788Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 6ba5b7ab-5472-4f04-ba3b-bd5d1c00b471 · outbound

This paper cites From Allies to Adversaries: Manipulating LLM Tool-Calling through Adversarial Injection.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems From Allies to Adversaries: Manipulating LLM Tool-Calling through Adversarial Injection

Reference 76

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.678358Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6fe0c16b-c9c5-4686-b103-ddb0ced94a13 · outbound

This paper cites Backdooralign: Mitigating fine-tuning based jailbreak attack with backdoor enhanced safety alignment.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Backdooralign: Mitigating fine-tuning based jailbreak attack with backdoor enhanced safety alignment

Reference 77

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:32:37.544261Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 1cd39cb2-ad4d-4097-9cd9-44ea269ef0ba · outbound

This paper cites A Comprehensive Survey in LLM(-Agent) Full Stack Safety: Data, Training and Deployment.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems A Comprehensive Survey in LLM(-Agent) Full Stack Safety: Data, Training and Deployment

Reference 78

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.685699Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation be9ee4a7-10a0-4f4b-a793-d24b6bec5b18 · outbound

This paper cites ScienceWorld: Is your Agent Smarter than a 5th Grader?.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems ScienceWorld: Is your Agent Smarter than a 5th Grader?

Reference 79

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.689405Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 785cfa32-51aa-4d6f-ad78-502d0157a6af · outbound

This paper cites G-Safeguard: A Topology-Guided Security Lens and Treatment on LLM-based Multi-agent Systems.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems G-Safeguard: A Topology-Guided Security Lens and Treatment on LLM-based Multi-agent Systems

Reference 80

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.692682Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b61d1d5a-cf4b-4993-a884-8647f3aef8da · outbound

This paper cites Augmenting language models with long-term memory.Advances in Neural Information Processing Systems, 36:74530–74543, 2023.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Augmenting language models with long-term memory.Advances in Neural Information Processing Systems, 36:74530–74543, 2023

Reference 81

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.695853Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c81ab498-e158-400f-84ac-8be0fcd0581e · outbound

This paper cites AgentGym: Evolving Large Language Model-based Agents across Diverse Environments.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems AgentGym: Evolving Large Language Model-based Agents across Diverse Environments

Reference 82

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.698999Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e48af40e-acf2-42d8-b9d9-dd4fc079521b · outbound

This paper cites The rise and potential of large language model based agents: A survey.Science China Information Sciences, 68(2):121101, 2025.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems The rise and potential of large language model based agents: A survey.Science China Information Sciences, 68(2):121101, 2025

Reference 83

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.702156Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f96d077d-7cfb-4ecc-949e-513f4ebb4f35 · outbound

This paper cites Certifiably robust rag against retrieval corruption.arXiv preprint arXiv:2405.15556, 2024.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Certifiably robust rag against retrieval corruption.arXiv preprint arXiv:2405.15556, 2024

Reference 84

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.704889Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 380e6eee-66c5-484b-86ee-9f201f1e5933 · outbound

This paper cites Bag of tricks: Benchmarking of jailbreak attacks on llms.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Bag of tricks: Benchmarking of jailbreak attacks on llms

Reference 85

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:32:37.521751Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 74828f52-df54-443a-8638-be1165b46fcd · outbound

This paper cites Qwen3 Technical Report.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Qwen3 Technical Report

Reference 86

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.710457Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 95703251-2882-404c-9a50-c7e497ae4a07 · outbound

This paper cites Gpt4tools: Teaching large language model to use tools via self-instruction.Advances in Neural Information Processing Systems, 36:71995–72007, 2023.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Gpt4tools: Teaching large language model to use tools via self-instruction.Advances in Neural Information Processing Systems, 36:71995–72007, 2023

Reference 87

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.713245Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a19666c5-0f5d-45e7-a5fb-c9349549add3 · outbound

This paper cites The second half.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems The second half

Reference 88

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:32:37.506455Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 7f61f150-b491-4527-9a47-b62e72b3d17d · outbound

This paper cites Webshop: Towards scalable real-world web interaction with grounded language agents.Advances in Neural Information Processing Systems, 35:20744–20757, 2022.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Webshop: Towards scalable real-world web interaction with grounded language agents.Advances in Neural Information Processing Systems, 35:20744–20757, 2022

Reference 89

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.719731Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fea6eceb-7aa5-4c7c-b05d-1c583cb0e5d6 · outbound

This paper cites On the vulnerability of safety alignment in open-access llms.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems On the vulnerability of safety alignment in open-access llms

Reference 90

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:32:37.490297Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation adf6ba7d-7317-400d-a0e8-27f41b4d2700 · outbound

This paper cites NetSafe: Exploring the Topological Safety of Multi-agent Networks.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems NetSafe: Exploring the Topological Safety of Multi-agent Networks

Reference 91

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.725488Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 658a354a-f690-49f2-985f-f7672cd019dc · outbound

This paper cites A Survey on Trustworthy LLM Agents: Threats and Countermeasures.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems A Survey on Trustworthy LLM Agents: Threats and Countermeasures

Reference 92

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.728262Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1524d6e6-286a-43ba-93ef-3b2fef453951 · outbound

This paper cites GPT-4 Is Too Smart To Be Safe: Stealthy Chat with LLMs via Cipher.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems GPT-4 Is Too Smart To Be Safe: Stealthy Chat with LLMs via Cipher

Reference 93

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.731379Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 00c62709-3317-40a3-b346-442a63fdae42 · outbound

This paper cites The Good and The Bad: Exploring Privacy Issues in Retrieval-Augmented Generation (RAG).

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems The Good and The Bad: Exploring Privacy Issues in Retrieval-Augmented Generation (RAG)

Reference 94

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.734755Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f589ac32-fc8d-4d1b-96e3-5b2e30bd4aa3 · outbound

This paper cites Breaking Agents: Compromising Autonomous LLM Agents Through Malfunction Amplification.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Breaking Agents: Compromising Autonomous LLM Agents Through Malfunction Amplification

Reference 95

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.737890Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ab6de568-c470-4aab-88e5-dea0578791ba · outbound

This paper cites Cut the Crap: An Economical Communication Pipeline for LLM-based Multi-Agent Systems.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Cut the Crap: An Economical Communication Pipeline for LLM-based Multi-Agent Systems

Reference 96

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.740954Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e4edd56d-f9ec-49a4-953c-f24c4b7be12c · outbound

This paper cites G-Designer: Architecting Multi-agent Communication Topologies via Graph Neural Networks.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems G-Designer: Architecting Multi-agent Communication Topologies via Graph Neural Networks

Reference 97

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.744658Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 957146b2-5704-4b00-aba6-c0e9972dae84 · outbound

This paper cites Multi-agent Architecture Search via Agentic Supernet.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Multi-agent Architecture Search via Agentic Supernet

Reference 98

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.747826Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 135758a2-101f-46b2-b1b3-b9f34e775ef4 · outbound

This paper cites On large language models safety, security, and privacy: A survey.Journal of Electronic Science and Technology, page 100301, 2025.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems On large language models safety, security, and privacy: A survey.Journal of Electronic Science and Technology, page 100301, 2025

Reference 99

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:32:37.480624Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 345cee46-9e13-41d1-b269-df216f2781e7 · outbound

This paper cites A Survey on the Memory Mechanism of Large Language Model based Agents.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems A Survey on the Memory Mechanism of Large Language Model based Agents

Reference 100

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.754225Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 076583d6-15c9-4396-a3d0-44469bf491fb · outbound

This paper cites Agent-SafetyBench: Evaluating the Safety of LLM Agents.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Agent-SafetyBench: Evaluating the Safety of LLM Agents

Reference 101

Resolution: unresolved; no resolver link, observed 2026-08-07T00:32:36.757251Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3a8ecb07-4fc3-4a68-ab6a-ff2bac3b1b42 · outbound

This paper cites Weak-to-strong jailbreaking on large language models.

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems Weak-to-strong jailbreaking on large language models

Reference 102

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:32:37.470778Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Pith citing papers

Observation 90e1e1ed-8764-4e66-a8d9-6960ea2c0b21 · inbound

A Large-Scale Evolvable Dataset for Model Context Protocol Ecosystem and Security Analysis cites this paper.

A Large-Scale Evolvable Dataset for Model Context Protocol Ecosystem and Security Analysis We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-06T21:44:54.864170Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cfe0e7e8-93f6-4771-ad9f-5e537f17e30b · inbound

A Survey of Context Engineering for Large Language Models cites this paper.

A Survey of Context Engineering for Large Language Models We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems

Reference 268

Resolution: verified exact; arxiv_id, observed 2026-05-13T20:58:45.454680Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation e0d13b95-ba3d-45c4-a254-a4f2d61dcdb2 · inbound

Quantifying Conversation Drift in MCP via Latent Polytope cites this paper.

Quantifying Conversation Drift in MCP via Latent Polytope We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-05T22:51:28.082763Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 20c9401e-5b28-422d-ba1a-9243586357e2 · inbound

SafeSearch: Automated Red-Teaming of LLM-Based Search Agents cites this paper.

SafeSearch: Automated Red-Teaming of LLM-Based Search Agents We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-04T14:43:49.514734Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fefa0c54-1a93-4642-b570-953c2be5843e · inbound

AgentBound: Securing Execution Boundaries of AI Agents cites this paper.

AgentBound: Securing Execution Boundaries of AI Agents We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems

Reference 11

Resolution: verified exact; arxiv_id, observed 2026-05-18T05:15:54.194195Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 3124087a-1487-43c7-acf0-c3166b8bb612 · inbound

BalDRO: A Distributionally Robust Optimization based Framework for Large Language Model Unlearning cites this paper.

BalDRO: A Distributionally Robust Optimization based Framework for Large Language Model Unlearning We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-03T10:45:41.323277Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 99d5a936-f809-49ba-9bd3-93cff40c539c · inbound

Combating Data Laundering in LLM Training cites this paper.

Combating Data Laundering in LLM Training We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems

Reference 16

Resolution: unresolved; no resolver link, observed 2026-07-13T14:08:35.474488Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 480a69b4-2b84-4ec8-a628-112915abefca · inbound

From Component Manipulation to System Compromise: Understanding and Detecting Malicious MCP Servers cites this paper.

From Component Manipulation to System Compromise: Understanding and Detecting Malicious MCP Servers We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems

Reference 16

Resolution: verified exact; arxiv_id, observed 2026-05-21T10:44:07.943736Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 90f557b0-dd05-47fa-bd65-952d9dc1467e · inbound

MCP-DPT: A Defense-Placement Taxonomy and Coverage Analysis for Model Context Protocol Security cites this paper.

MCP-DPT: A Defense-Placement Taxonomy and Coverage Analysis for Model Context Protocol Security We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems

Reference 10

Resolution: verified exact; arxiv_id, observed 2026-05-10T21:10:46.722140Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 6bfb7f42-7003-402d-9b36-9ebc80aaad4c · inbound

"What Happens Locally, Leaks Globally": Detecting Privacy Leakage Risks in MCP Servers cites this paper.

"What Happens Locally, Leaks Globally": Detecting Privacy Leakage Risks in MCP Servers We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems

Reference 9

Resolution: metadata mismatch; arxiv_id, observed 2026-07-04T06:49:38.283917Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.