🧰 Model Context Protocol

The items that name it 359

• Tanmay (@TanmayK_03) on X
  threadas classification.technologies2026-10-30
• Rohan Paul (@rohanpaul_ai) on X
  threadas classification.technologies2026-10-29
• The Rise of Agentic Attacks: What They Look Like, and Where They're Headed
  articleas affected products2026-10-29It is written here as MCP servers.
• The Rise of Agentic Attacks: What They Look Like, and Where They're Headed
  writeupas affected products2026-10-29It is written here as MCP servers (stdio-transport RCE).

  stdio-transport RCE

• Cisco Open Sources Quarry to Catch AI Agent Attacks Hidden in Harmless Steps
  articleas classification.technologies2026-10-29
• What is an AIBOM? AI bill of materials explained for security teams | Endor Labs
  articleas classification.technologies2026-10-29
• ADLC metrics: How to measure agentic development security | Endor Labs
  articleas classification.technologies2026-10-29
• PCM: A Proof-Carrying Message Protocol for Verifiable Multi-Agent Communication
  paperas classification.technologies2026-10-29
• Rohan Paul (@rohanpaul_ai) on X
  threadas classification.technologies2026-10-28
• Cryptography and Security Papers (@FSFG) on X
  paperas classification.technologies2026-10-28
• Hakimi Eiqbal (@hakimieiqbal) on X
  threadas classification.technologies2026-10-28
• ADLC compliance: Mapping agentic development to NIST SSDF, OWASP, and the EU AI Act | Endor Labs
  articleas classification.technologies2026-10-28
• The ADLC maturity model: 5 stages of agentic development security | Endor Labs
  articleas classification.technologies2026-10-28
• How to evaluate ADLC security solutions: A buyer's checklist | Endor Labs
  articleas classification.technologies2026-10-28
• Kizuno18 (@Kizuno18) on X
  threadas classification.technologies2026-10-27
• Buchi Reddy B (@buchireddy) on X
  threadas classification.technologies2026-10-27
• cubillas (@cubillas888) on X
  threadas classification.technologies2026-10-27
• Nicholas Martin (@ndgmlondon) on X
  threadas classification.technologies2026-10-27
• Measuring Users' Security and Privacy Concerns About Third-Party Community Bots on Discord
  paperas classification.technologies2026-10-27
• Log-To-Leak: Prompt Injection Attacks on Tool-Using LLM Agents via the Model Context Protocol
  paperas classification.technologies2026-10-26
• David Gutierrez | Eng (@david_x_ai) on X
  threadas classification.technologies2026-10-26
• Benjamin👨‍💻 (@Hislordship_9) on X
  threadas classification.technologies2026-10-26
• David Gutierrez | Eng (@david_x_ai) on X
  threadas classification.technologies2026-10-25
• Dan Kornas (@DanKornas) on X
  toolas classification.technologies2026-10-25
• Deletion, Substitution, Repair and Transplant: Component Attribution in BARBICAN, a Layered Prompt-Injection Gateway for MCP
  paperas classification.technologies2026-10-23
• ADLC Security Best Practices for Teams Shipping with Coding Agents | Endor Labs
  articleas classification.technologies2026-10-23
• Yash Patil (@yvshpvtil) on X
  threadas classification.technologies2026-10-23
• Agentic AI Defense: Decode-Time Provenance Binding for MCP
  paperas classification.technologies2026-10-22
• VibecodersHQ (@VibeCodrHQ) on X
  threadas classification.technologies2026-10-21
• Swif (@swif_ai) on X
  threadas classification.technologies2026-10-21
• Software supply chain security best practices for 2026
  articleas classification.technologies2026-10-21
• AI Model Risk Depends on What the Model Can Reach
  articleas classification.technologies2026-10-21
• Combining Long-Term Memory and LLMs Comes at a High Cost: From Prompt Injection to Persistent Compromise
  paperas classification.technologies2026-10-20
• DFIR Radar (@DFIR_Radar) on X
  threadas classification.technologies2026-10-19
• Huda (@AlAssaf_H) on X
  articleas classification.technologies2026-10-19
• Panos Gkatziroulis 🦄 (@ipurple) on X
  repoas classification.technologies2026-10-19
• Santhosh Patel (@santhosh_patell) on X
  threadas classification.technologies2026-10-19
• FastFix AI (@FastfixAI) on X
  threadas classification.technologies2026-10-18
• Magai (@HeyMagai) on X
  threadas classification.technologies2026-10-18
• Steven Daugherty (@StevenDaugherty) on X
  threadas classification.technologies2026-10-17
• Mikko Laakkonen (@mikkolaakkonen) on X
  threadas classification.technologies2026-10-17
• Bravish (@bravish_) on X
  toolas classification.technologies2026-10-17
• Nicholas Martin (@ndgmlondon) on X
  threadas classification.technologies2026-10-16
• In Other News: ICS-Sniper Attack, Hacking XWorm Malware Operators, $250k Chrome Exploit
  articleas classification.technologies2026-10-16
• P.K. Sharma (@_pksharma) on X
  threadas classification.technologies2026-10-15
• My Credentials, Please: Detecting Filesystem Capability Laundering in MCP Servers at Scale
  paperas classification.technologies2026-10-15
• [tl;dr sec] #350 - Reverse Engineer Anything, Tracking AI-integrated Malware, Test-driving Jev
  articleas classification.technologies2026-10-15
• Francis Santerre (@santerref) on X
  threadas classification.technologies2026-10-15
• dbugs (@ptdbugs) on X
  toolas classification.technologies2026-10-15
• Approve What, Exactly? Effect Binding and Trusted Rendering for Agent Step-Up Authorization
  paperas classification.technologies2026-10-15
• Thinking in Graphs: Neo4j-Backed Exploit-Chain Reasoning for LLM Penetration-Testing Agents
  paperas tools2026-10-14It is written here as Model Context Protocol (MCP) (connects the agent to the graph).

  MCP

• Victor (@VMG_CU) on X
  threadas classification.technologies2026-10-14
• 44er (@44errrr) on X
  toolas classification.technologies2026-10-14
• Visual Studio Copilot and SSMS Remote Code Execution (CVE-2026-103572)
  writeupas classification.technologies2026-10-13
• Phillip Shoemaker (@pbsIdentity) on X
  threadas classification.technologies2026-10-12
• Governing the Agentic SDLC Through Continuous Policy Enforcement
  articleas classification.technologies2026-10-12
• State of Dependency Management 2025 | Endor Labs
  paperas classification.technologies2026-10-12
• The OWASP Top 10 Gets Modernized | Endor Labs
  articleas classification.technologies2026-10-12
• Context Engineering for Application Security | Endor Labs
  articleas classification.technologies2026-10-12
• Nicholas Michael Kloster (@showxlate) on X
  threadas classification.technologies2026-10-12
• Dan Kornas (@DanKornas) on X
  repoas classification.technologies2026-10-11
• KedgeFlow: A Programmable Control Plane for Enterprise AI Agents
  paperas tools2026-10-10It is written here as Model Context Protocol (MCP).

  MCP

• morluto/awesome-reverse-engineering
  repoas classification.technologies2026-10-09
• Richard Abishai (@richardabish_ai) on X
  threadas classification.technologies2026-10-09
• trhhubsd/ARTEX
  repoas classification.technologies2026-10-08
• James Berthoty (@JamesBerthoty) on X
  threadas classification.technologies2026-10-08
• SonicWall and Splunk Patch Critical Vulnerabilities
  articleas classification.technologies2026-10-08
• Top MCP security resources — October 2026
  articleas classification.technologies2026-10-08
• Applying Security by Design at the Point of Execution: How Governed Security Requirements Affect the Security of AI-Generated Code
  paperas classification.technologies2026-10-07
• Identity-aware AI data agents with AWS Lake Formation and Trusted Identity Propagation
  articleas classification.technologies2026-10-06
• Prasenjit Sarkar (@stretchcloud) on X
  threadas classification.technologies2026-10-06
• Jan Curn (@jancurn) on X
  threadas classification.technologies2026-10-03
• COPEX: Benchmarking LLM Robustness to Adversarial Context Across Model Context Protocol Layers
  paperas classification.technologies2026-10-03
• Nation (@visitnation) on X
  threadas classification.technologies2026-10-03
• CyTReX: Explainable AI-Based Cybersecurity Threat Reasoning Framework for DER Networks
  paperas tools2026-10-03It is written here as Model Context Protocol (MCP).

  Enables cyber threat intelligence (CTI) enrichment integrated into the evidence packet

• Six defects in one signature verification tool, found from outside in six rounds
  writeupas classification.technologies2026-10-02
• Evaluating and Improving the Robustness of Large Language Models to Input Sequence Variations
  paperas classification.technologies2026-10-01
• Evaluating and Improving the Robustness of Large Language Models to Input Sequence Variations
  paperas classification.technologies2026-10-01
• Identity Management for Agentic AI [pdf] (2025)
  paperas tools2026-10-01It is written here as Model Context Protocol (MCP).

  standardized protocol connecting language model-based agents to external data sources and tools; noted as leading in adoption but initially lacked authentication

• Average AI Bro (@AverageAiBro) on X
  toolas classification.technologies2026-09-30
• Connect Orca’s ChatGPT Plugin: Cloud Risk Context in Chat and Codex
  articleas classification.technologies2026-09-30
• Vulnerability Discovery and Exploitation Trends in the AI Era (Google Cloud)
  articleas classification.technologies2026-09-30
• Getting the Source Right, Not Just the Fact: Source-Aware Verification for MCP Agents
  articleas classification.technologies2026-09-29
• OperTraitors: How Kubernetes Operators Betray Your Security Posture
  articleas tools2026-09-29

  Example channel for external agent bridges that connect external AI agents to operators

• graygnatconsole/mcp-audit-tool
  repoas classification.technologies2026-09-26
• Proactive Defense: Hardening Code Pipelines and CI/CD Infrastructure
  articleas classification.key entities2026-09-24
• Static Defenses vs. Dynamic Exploits: Why A2AS Fails at the Runtime
  writeupas classification.technologies2026-09-23
• Top 10 attacks on Claude Code and what each one actually broke
  writeupas classification.technologies2026-09-23
• 30+ MCP CVEs in Two Months: What the Early-2026 Vulnerability Wave Means for Your MCP Supply Chain
  articleas classification.technologies2026-09-23
• A2M: Trace-Optimized Agent Hijacking in the MCP Ecosystem
  paperas affected products2026-09-22It is written here as Model Context Protocol (MCP).

  Agent ecosystem relying on semantic matching to select third-party tools, exposed to semantic supply-chain risk via attacker-controlled metadata and tool outputs

• npm Supply-Chain Attack Abuses Trusted Publishing to Ship GHAPPIER Loader
  articleas classification.technologies2026-09-22
• Shor
  repoas classification.technologies2026-09-19
• Outrider Recon
  repoas classification.technologies2026-09-19
• Zero-Trust Authorization and Discovery for Enterprise MCP
  paperas tools2026-09-18It is written here as Model Context Protocol (MCP).

  Widely adopted interface/protocol for LLM agent tool-call authorization boundary

• VIPER-MCP: Detecting Vulnerabilities in MCP Servers
  paperas classification.technologies2026-09-17
• AI Agent Hacking Writeups
  repoas affected products2026-09-17It is written here as Model Context Protocol (MCP).

  MCP

• Closed-World Resolution Against Tool Hallucination in LLM Agents
  paperas affected products2026-09-16It is written here as Model Context Protocol (MCP).

  Merging several servers into one namespace creates hallucination surfaces (collisions and shadowing) a single registry cannot express

• What is an agent harness (and how do you secure one)?
  articleas classification.technologies2026-09-14
• VIPER-MCP paper
  paperas classification.technologies2026-09-13
• shinthink/blitzstrike
  repoas classification.technologies2026-09-12
• Anti-Pattern Avoidance: A Prompt Pattern for Safer AI-Generated Code | Blog | Endor Labs
  articleas tools2026-09-11It is written here as MCP servers.

  provide real-time security context and guardrails alongside secure prompting

• Off Guard: Breaking LiteLLM from authentication bypass to cloud compromise
  writeupas classification.technologies2026-09-09
• 0xor0ne (@0xor0ne) on X
  threadas classification.technologies2026-09-09
• [tl;dr sec] #344 - VMs won't contain Cyber-capable Agents, AWS AI Security Analyst, Decompilers vs LLMs
  articleas classification.technologies2026-09-08
• PlamenTSV/plamen
  repoas classification.technologies2026-09-07
• 0x4D31/endpoint-ai-agent-abuse
  repoas classification.technologies2026-09-07
• mukul975/cve-mcp-server
  repoas classification.technologies2026-09-07
• Top MCP Security Resources — September 2026 (Adversa.ai)
  articleas classification.technologies2026-09-07
• Top Agentic AI security resources — September 2026
  articleas classification.technologies2026-09-04
• Grafana fixes critical SSRF flaw affecting Grafana MCP servers
  advisoryas classification.technologies2026-09-03
• Agentic Development Security: How to Secure Code Your Agents Write | Blog | Endor Labs
  articleas classification.technologies2026-09-02
• The Secure Agentic Development Lifecycle (ADLC), Explained | Blog | Endor Labs
  articleas classification.technologies2026-09-02
• Secrets in AI-Generated Code: How Coding Agents Leak Credentials | Blog | Endor Labs
  articleas classification.technologies2026-09-02
• Prompt Injection Against Coding Agents: The Attack Surface Nobody Owns | Blog | Endor Labs
  articleas classification.technologies2026-09-02
• The Evolution of JFrog AI Catalog: Your AI Control Plane for Agentic Development
  articleas classification.technologies2026-09-02
• An AI-Assisted Cyber Attack: Inside a Unit 42 Investigation
  writeupas tools2026-09-02It is written here as Model Context Protocol (MCP).

  Referenced as an AI tool integration/gateway that organizations should inventory and govern as core infrastructure

• Hooks are the control plane for the agentic development lifecycle | Blog | Endor Labs
  articleas classification.technologies2026-09-01
• Lose Control Flow: Unauthenticated Tool Execution in Dolt MCP
  advisoryas classification.technologies2026-09-01
• Awesome Cybersecurity Agentic AI
  repoas classification.technologies2026-08-31
• LiteLLM
  toolas classification.technologies2026-08-31
• UI-TARS-desktop MCP Server: Critical Unauthenticated RCE
  advisoryas classification.technologies2026-08-27
• Inside 90 days of attacks on AI infrastructure
  writeupas classification.technologies2026-08-27
• [tl;dr sec] #343 - Hugging Face Technical Report, AWSHound, OWASP Agentic Skills Top 10
  articleas classification.technologies2026-08-27
• IAM Policy Autopilot
  repoas classification.technologies2026-08-25
• Cloudflare Computer
  repoas classification.technologies2026-08-25
• Cybermes
  repoas classification.technologies2026-08-23
• duty1g/x64dbg-mcp-server
  repoas classification.technologies2026-08-22
• The New MCP Roadmap
  articleas affected products2026-08-22It is written here as Model Context Protocol (MCP).

  Protocol specification; 2026-07-28 release removed protocol-level sessions and initialization handshake (SEP-2575, SEP-2567); remote MCP servers now operate as standard HTTP workloads

• crt.name
  toolas classification.technologies2026-08-21
• JFrog: Propagating User Identity From AI Agents to Your Tools
  articleas classification.technologies2026-08-21
• Uber ADR (Agentic AI Detection and Response)
  repoas classification.technologies2026-08-21
• AI Development and Agentic Security Labs
  courseas classification.technologies2026-08-21
• N4D Mesh Controller: New infrastructure, a UPX-packed agent labeled "go-titan," and how to hunt for it
  writeupas classification.technologies2026-08-20
• GhostSplice: Malicious MCP Servers Split Exfiltration to Bypass Guardrails
  articleas affected products2026-08-19It is written here as Model Context Protocol (MCP).

  Open standard letting AI assistants call external tools; targeted protocol for the GhostSplice attack.

• 0sec-labs/0
  repoas classification.technologies2026-08-19
• Remediation Agents, Demystified: Why Fixing Beats Finding
  articleas classification.technologies2026-08-19
• Socket: How AI Agents Expand the Software Supply Chain Attack Surface
  talkas classification.technologies2026-08-17
• What is an agent harness? The software that turns a model into an agent | Blog | Endor Labs
  articleas classification.technologies2026-08-15
• The OpenSourceMalware Show #17
  talkas classification.technologies2026-08-14
• Orca Security: Shadow AI Governance
  articleas classification.technologies2026-08-13
• The Agent Baseline: 35 Controls, But Where Should You Start?
  articleas classification.technologies2026-08-12
• xiaoyeuzixi/revlab
  repoas classification.technologies2026-08-11
• GhostJacking Attacks: AI Agent Takeover via Firewall Evasion
  writeupas classification.technologies2026-08-10
• AuditSentry
  repoas classification.technologies2026-08-10
• Cloud Threat Highlights: H1 2026
  writeupas classification.technologies2026-08-06
• Top MCP security resources — August 2026
  articleas classification.technologies2026-08-06
• Can AI Do Novel Security Research? Meet the HTTP Terminator
  writeupas classification.technologies2026-08-05
• Can AI do novel security research? Meet the HTTP Terminator (James Kettle)
  writeupas classification.technologies2026-08-05
• Dark-Moon
  repoas classification.technologies2026-08-05
• SecureAI-Scan
  repoas classification.technologies2026-08-05
• Top AI Coding Agent security resources — August 2026
  articleas classification.technologies2026-08-03
• Before the first prompt: Code execution paths in trusted coding-agent projects
  writeupas tools2026-08-03It is written here as Model Context Protocol (MCP).

  Local stdio MCP servers defined in project config are processes the agent starts, which gave code execution on project open in Codex.

• Nomos
  repoas classification.technologies2026-08-02
• Awesome Offensive AI Agentic Landscape
  repoas classification.technologies2026-08-01
• AgentHound
  repoas classification.technologies2026-08-01
• Introducing AI Coding Agent Governance: Using Hooks to Bring Visibility to the ADLC | Blog | Endor Labs
  articleas classification.technologies2026-07-31
• Critical Security Controls for Governing AI Coding Agents | Blog | Endor Labs
  articleas classification.technologies2026-07-31
• Top Agentic AI security resources — August 2026
  articleas classification.technologies2026-07-31
• OpenOSINT
  repoas classification.technologies2026-07-31
• Rapid7
  articleas classification.technologies2026-07-30
• Claude as Autonomous Bug Bounty Hunter
  articleas classification.technologies2026-07-30
• Slopsquatting: When AI Agents Hallucinate Malicious Packages | Blog | Endor Labs
  articleas classification.technologies2026-07-29
• PentesterFlow/agent
  repoas classification.technologies2026-07-29
• Bug Bounty Directory & MCP API
  repoas classification.technologies2026-07-29
• RamanMG/Burp-MCP-Unrestricted
  repoas tools2026-07-29It is written here as Model Context Protocol (MCP).

  Protocol used to connect Burp Suite to AI clients such as Claude Desktop

• HCSEC-2026-23 - Multiple vulnerabilities impacting HashiCorp Terraform MCP Server
  advisoryas classification.technologies2026-07-28
• The risk hiding behind exposed MCP servers
  writeupas classification.technologies2026-07-28
• The 2026-07-28 MCP Specification: Stateless Core, MRTR, Header-Based Routing
  articleas tools2026-07-28It is written here as Model Context Protocol (MCP).

  Data and interactivity substrate for agentic workflows; protocol under discussion

• Dalfox
  repoas classification.technologies2026-07-25
• The OpenSourceMalware Show #14
  talkas classification.technologies2026-07-23
• MCP Configuration Poisoning: Owning Your Machine With Just A Text File
  writeupas classification.technologies2026-07-23
• BountyForge
  repoas classification.technologies2026-07-22
• Beyond MCP: The New Security Playbook for Coding Agents | Blog | Endor Labs
  articleas classification.technologies2026-07-21
• When Your AI Reviewer Works for the Attacker: A Confused-Deputy Bug in Microsoft
  writeupas classification.technologies2026-07-21
• Security notice and deprecation: standalone circleci-mcp-server
  advisoryas classification.technologies2026-07-20
• AWS Kiro RCE: Prompt Injection to Code Execution | Kodem
  writeupas classification.technologies2026-07-19
• AI Security Briefing
  articleas classification.technologies2026-07-18
• AI Security Briefing
  writeupas classification.technologies2026-07-18
• AI Security Briefing
  articleas classification.technologies2026-07-18
• Awesome AI Web3 Security
  repoas classification.technologies2026-07-18
• AI-Assisted Vulnerability Management
  articleas classification.technologies2026-07-18
• jit
  repoas classification.technologies2026-07-18
• CVE-2026-59950
  advisoryas classification.technologies2026-07-17
• AI Security Briefing
  advisoryas classification.technologies2026-07-17
• Solving GitHub's Secure Code game with an AI red teaming agent
  writeupas classification.technologies2026-07-17
• CVE-2026-55608
  advisoryas classification.technologies2026-07-16
• Strix
  repoas tools2026-07-15It is written here as Model Context Protocol (MCP).

  Protocol Strix can connect to (via ~/.strix/mcp-servers.json) to expose external server tools (e.g. GitHub) to the agent during a run

• Skills Janitor
  repoas classification.technologies2026-07-15
• HOL Guard
  repoas classification.technologies2026-07-15
• brain0
  repoas classification.technologies2026-07-15
• The Cursor deeplink vulnerability that turns a “review this PR” click into remote code execution
  writeupas classification.technologies2026-07-15
• JFrog: Managing AI Agent Primitives with APM
  articleas classification.technologies2026-07-15
• deep-scan
  repoas classification.technologies2026-07-14
• WSO2 Trusted AI Governance
  paperas classification.technologies2026-07-14
• deep scan tuto
  writeupas classification.technologies2026-07-14
• What is a dependency firewall?
  articleas classification.technologies2026-07-13
• AI Security Briefing
  advisoryas classification.technologies2026-07-13
• CVE-2026-61459
  advisoryas classification.technologies2026-07-12
• awesome-ai-security-tools
  repoas classification.technologies2026-07-11
• apisec-inc/AI-Surface
  repoas classification.technologies2026-07-11
• requie/AI-Red-Teaming-Guide
  repoas classification.technologies2026-07-08
• elder-plinius/T3MP3ST
  repoas classification.technologies2026-07-07
• One Request to Own Every Repo: How We Hijacked GitLab Through Its MCP Server - Pluto Security
  writeupas classification.technologies2026-07-06
• AI Security Briefing
  advisoryas classification.technologies2026-07-05
• AI Security Briefing
  advisoryas classification.technologies2026-07-05
• CVE-2026-14699: Symlink Following in zcaceres markdownify-mcp - Live Threat Intelligence - Threat Radar | OffSeq.com
  advisoryas classification.technologies2026-07-05
• GitHub Secure Code Game
  courseas classification.technologies2026-07-04
• prempti
  repoas classification.technologies2026-07-04
• ThinkWatch
  repoas classification.technologies2026-07-04
• 0xSteph/pentest-ai
  repoas classification.technologies2026-07-04
• Advanced Prompt Injection Techniques 2026: 7 Attack Chains
  articleas tools2026-07-04It is written here as Model Context Protocol (MCP).

  Standard protocol for connecting LLM systems to external tools; installed in Cursor, Claude Desktop, VS Code Copilot; has documented tool poisoning, rug-pull, and cross-server shadowing vulnerabilities.

• Vulnerability-Lookup
  advisoryas classification.technologies2026-06-29
• MCP Auto-Execution: From Git Clone to Cloud Compromise in Amazon Q VS Code Extension
  writeupas tools2026-06-26It is written here as Model Context Protocol (MCP).

  Standard allowing AI assistants to spawn local processes/MCP servers and interact with external services (databases, APIs, build tools, local resources)

• Incident Report: CVE-2026-LGTM
  articleas classification.technologies2026-06-26
• What nearly 10,000 developer environments reveal about agentic development risk
  articleas classification.technologies2026-06-23
• 8kSec AI/LLM Exploitation Labs
  courseas classification.technologies2026-06-21
• Teycir/SkillsGuard
  repoas classification.technologies2026-06-21
• Armur-Ai/Pentest-Swarm-AI
  repoas classification.technologies2026-06-21
• AutoJack: How a single page can RCE the host running your AI agent | Microsoft Security Blog
  writeupas tools2026-06-19It is written here as Model Context Protocol (MCP).

  Protocol for connecting agents to tool/server processes; the vulnerable WebSocket control plane

• The OpenSourceMalware Show #9
  talkas classification.technologies2026-06-18
• Vulnerability-Lookup
  advisoryas classification.technologies2026-06-18
• One Fake Bug Report Hijacked a $250B Company’s AI Agent
  articleas classification.technologies2026-06-17
• Pipelock
  repoas classification.technologies2026-06-14
• SkillSpector
  repoas classification.technologies2026-06-14
• Endor Labs: AI AppSec Agents Cost 12x More Without Pre-Computed Security Context
  articleas tools2026-06-11It is written here as Model Context Protocol (MCP).

  Anthropic's tool-use architecture; cited as independent work showing token reductions up to 98% with accuracy preserved or improved

• Hacking Google with A.I. for $500,000
  writeupas classification.technologies2026-06-11
• ShulkwiSEC/bb-huge
  repoas classification.technologies2026-06-05
• Updating the taxonomy of failure modes in agentic AI systems: What a year of red teaming taught us | Microsoft Security Blog
  articleas tools2026-06-04It is written here as Model Context Protocol (MCP).

  De facto standard protocol for connecting AI models to external tools; source of tool poisoning and other vulnerabilities

• Claude.ai Prompt Injection Vulnerability | Oasis Security
  writeupas classification.technologies2026-06-03
• What MDM can't protect on developer machines (and what to do about it)
  articleas classification.technologies2026-05-28
• mukul975/Anthropic-Cybersecurity-Skills
  repoas classification.technologies2026-05-26
• AWS MCP Server (GA)
  articleas classification.technologies2026-05-24
• Bumblebee
  repoas classification.technologies2026-05-24
• CyberStrikeus/CyberStrike
  repoas classification.technologies2026-05-23
• elementalsouls/Claude-BugHunter
  repoas classification.technologies2026-05-22
• The 2026-07-28 MCP Specification Release Candidate
  articleas classification.technologies2026-05-21
• JFrog: The Governance Gap Between Your Policy and Your Pipeline
  articleas classification.technologies2026-05-21
• JFrog: Keep your Agents Under Control with agent-belt
  toolas classification.technologies2026-05-19
• H-mmer/pentest-agents
  repoas classification.technologies2026-05-18
• Open ASM Platform
  repoas classification.technologies2026-05-17
• FoxGuard
  repoas classification.technologies2026-05-16
• i8void/cybersleuth
  repoas classification.technologies2026-05-14
• Security Architecture of GitHub Agentic Workflows
  articleas classification.technologies2026-05-13
• Bullying LLMs
  writeupas tools2026-05-13It is written here as Model Context Protocol (MCP).

  Protocol exposing tools to Claude Code as callable typed functions; used to build 8 MCP servers across 5 VMs with 300+ tools

• GigaChad GRC
  repoas tools2026-05-13It is written here as MCP servers (three repository MCP servers).

  Packaged and launched by default Controls image; require their own credentials and access to upstream tools they call

• CSA: VulnOps in the Age of AI
  articleas classification.technologies2026-05-13
• PromptZero
  repoas classification.technologies2026-05-12
• paralax/SeldomMonster
  repoas classification.technologies2026-05-12
• matty69v/Bug-Bounty-Agents
  repoas classification.technologies2026-05-12
• GitHub Secret Scanning MCP Server
  articleas classification.technologies2026-05-11
• GitHub Dependency Scanning MCP Server
  toolas classification.technologies2026-05-11
• Pantheon-Security/medusa
  repoas classification.technologies2026-05-10
• GH05TCREW/pentestagent
  repoas classification.technologies2026-05-08
• 0xhackerfren/ProcMon-MCP
  repoas classification.technologies2026-05-07
• anmolksachan/AI-ML-Free-Resources-for-Security-and-Prompt-Injection
  repoas classification.technologies2026-05-07
• TrustFall: 1-click coding agent RCE in Claude, Cursor, Copilot
  writeupas classification.technologies2026-05-07
• IronCurtain: Secure AI Agent Runtime
  repoas classification.technologies2026-05-07
• GigaChad GRC
  repoas tools2026-05-06It is written here as MCP servers (three repository servers).

  Packaged and launched by default Controls image; require credentials for any upstream tools they call

• MCP Audit Extension
  repoas classification.technologies2026-05-06
• Zero-Click RCE in MCP
  writeupas tools2026-05-06It is written here as Model Context Protocol (MCP).

  Protocol allowing LLMs/agents to call external tools; the integration mechanism abused in the attack

• Trail of Bits AI-Native Journey
  articleas classification.technologies2026-05-06
• h1-brain
  repoas classification.technologies2026-05-06
• mcpwn
  repoas classification.technologies2026-05-06
• Microsoft Agent Governance Toolkit
  repoas classification.technologies2026-05-06
• BishopFox/aimap
  repoas classification.technologies2026-05-06
• Octelium
  repoas classification.technologies2026-05-05
• requie/LLMSecurityGuide
  repoas classification.technologies2026-05-05
• PatrikFehrenbach/h1-brain
  repoas classification.technologies2026-05-05
• GitHub RCE Vulnerability: CVE-2026-3854 Breakdown | Wiz Blog
  writeupas classification.technologies2026-04-28
• MCP TypeScript SDK Cross-Client Data Leak (CVE-2026-25536)
  advisoryas classification.technologies2026-02-04
• MCPJam Inspector RCE (CVE-2026-23744)
  advisoryas classification.technologies2026-02-01
• Claude Cowork Exfiltrates Files
  writeupas tools2026-01-27It is written here as MCP servers.

  Connected integrations cited as an untrusted data source and risk surface for prompt injection

• rikugan: LLM reverse-engineering agent for IDA Pro and Binary Ninja
  repoas classification.technologies
• VulnCheck - Outpace Adversaries
  toolas tools

  Protocol implemented by the VulnCheck MCP server for receiving vulnerability and exploit intelligence

• SAIL Framework: A Practical Guide for AI Security
  articleas classification.technologies
• Search Search results for "security" | MCP Market
  otheras classification.technologies
• better-isms/grcskills
  repoas classification.technologies
• ToFu: Exploiting and Fuzzing the LLM–Orchestrator–Tool Trust Boundary
  paperas classification.technologies
• OWASP/MCP-Security-Testing-Guide
  repoas classification.technologies
• morluto/rea
  repoas classification.technologies
• santosomar/AI-agents-for-cybersecurity
  repoas classification.technologies
• AI Agents for Cybersecurity: A Guided Tour of a Repository with Tons of Resources and Examples
  articleas classification.technologies
• AI Agents for Cybersecurity: A Guided Tour of a Repository with Tons of Resources and Examples
  articleas classification.technologies
• bilawalsidhu/gods-eye-view
  repoas classification.technologies
• https://media.licdn.com/dms/document/media/v2/D4D1FAQGtvK5tUnN0EQ/feedshare-document-url-metadata-scrapper-pdf/B4DaEgDW2kKUA4-/0/1791539116721?e=1792170000&v=beta&t=6lNyWkJparrDrHu4bE6lW0AGmrORP_M9VmFbuMKjk98
  courseas classification.technologies
• ETDI: Comprehensive MCP Security Framework
  writeupas affected productsIt is written here as Model Context Protocol (MCP).

  Standard specification lacks security primitives for tool authenticity and integrity, exposing LLM applications to Tool Poisoning and Rug Pull attacks

• The Vulnerable MCP Project — MCP vulnerability database
  repoas classification.technologies
• AI-Infra-Guard (A.I.G) — Tencent Zhuque Lab
  repoas classification.technologies
• openrig — multi-agent Claude Code/Codex orchestration
  repoas classification.technologies
• MCP Python SDK v2.2.0
  otheras classification.technologies
• Delegation Without Trust: An Empirical Gap Analysis of Identity, Authorization, and Runtime Governance in Multi-Agent LLM Systems
  paperas classification.technologies
• Kong Konnect MCP server: stored prompt injection via analytics data leads to credential/data exfiltration (CVE-2026-13341)
  advisoryas classification.technologies
• Agent Threat Rules (ATR) — open detection-rule standard for AI agent threats
  repoas classification.technologies
• Progressive Skill Discovery as Access Control for Tool-Using LLM Agents
  paperas classification.technologies
• Agent Threat Rules (ATR) — Sigma/YARA-style open detection format for AI agent threats
  otheras classification.technologies
• CVE-2026-90617 in PentestAgent
  advisoryas classification.technologies
• CVE-2026-85787 in postgres-mcp-server
  advisoryas classification.technologies
• CVE-2026-85661 in excel-mcp-server
  advisoryas toolsIt is written here as Model Context Protocol (MCP).

  MCP

• Veeam, Terraform MCP, Django Patch Critical Flaws, Led by CVSS 10.0 Cross-Tenant Bug
  articleas classification.technologies
• Marimo Notebook Flaw Could Run MCP Commands Before Cells Execute in Edit Mode
  articleas classification.technologies
• Amazon Kiro Prompt Injection Can Exfiltrate Sensitive Data Through Kiro Powers
  articleas toolsIt is written here as Model Context Protocol (MCP).

  Protocol used to configure server connections and tools for AI agents, implicated in multiple vulnerabilities

• FakeGit Campaign Uses 7,600 GitHub Repositories to Spread SmartLoader Malware
  articleas classification.technologies
• Critical Cursor Flaws Could Let Prompt Injection Escape Sandbox and Run Commands
  articleas toolsIt is written here as Model Context Protocol (MCP).

  Protocol connecting Cursor's agent to external services (e.g. Linear workspace); used as an injection vector

• LiteLLM Flaw CVE-2026-42271 Exploited in the Wild, Chains to Unauthenticated RCE
  articleas classification.technologies
• AutoJack Attack Lets One Web Page Hijack AI Agent for Host Code Execution
  articleas classification.technologies
• Agentjacking Attack Tricks AI Coding Agents Into Running Malicious Code
  articleas toolsIt is written here as Model Context Protocol (MCP).

  Protocol used by AI agents to connect to external services like Sentry; implicit trust in MCP responses enables the attack

• 973 MCP Packages, 71% Single-Maintainer: A Practitioner
  articleas classification.technologies
• Vulnerability Database - OSV
  advisoryas classification.technologies
• CVE-2026-75130: Context7 through 2
  advisoryas classification.technologies
• CVE-2026-65695: Office-Word-MCP-Server through 1
  advisoryas classification.technologies
• CVE-2026-65056: mcp-webresearch 0
  advisoryas classification.technologies
• CVE-2026-57860: ForgeCode (tailcallhq/forgecode), an AI pair-programming CLI, automatically loads and executes the MCP servers defined i
  advisoryas classification.technologies
• CVE-2026-55544: NextCRM is open-source customer relationship management (CRM) software
  advisoryas classification.technologies
• CVE-2026-52869: The MCP Python SDK, called mcp on PyPI, is a Python implementation of the Model Context Protocol (MCP)
  advisoryas classification.technologies
• CVE-2026-49291: mcp-memory-service is a semantic memory layer for AI applications
  advisoryas classification.technologies
• CVE-2026-42271: LiteLLM is a proxy server (AI Gateway) to call LLM APIs in OpenAI (or native) format
  advisoryas classification.technologies
• CVE-2026-19984: A flaw has been found in jkawamoto mcp-florence2 up to 0
  advisoryas classification.technologies
• CVE-2026-19978: A flaw has been found in jiantao88 android-mcp-server up to cfb872b2446794193b58edd63f4dbf6af48a6292
  advisoryas classification.technologies
• CVE-2026-18655: Improper restriction of intended endpoints in the RabbitMQ broker connection tools of the Amazon MQ MCP Server (awslabs
  advisoryas classification.technologies
• CVE-2026-16584: Improper handling of an initialization failure in AWS API MCP Server from 0
  advisoryas classification.technologies
• CVE-2026-16498: The terraform-mcp-server before version 1
  advisoryas classification.technologies
• CVE-2026-14748: A flaw has been found in AIAnytime Awesome-MCP-Server up to a884bb51bcd99e08e14fd712c749d55d9d9a13ab
  advisoryas classification.technologies
• CVE-2026-14541: An authentication bypass and audience confusion vulnerability exists in the Google OAuth provider component of Google mc
  advisoryas classification.technologies
• Unauthenticated Administrator takeover exfiltrated via the MCP
  advisoryas classification.technologies
• "Allowed HTTP Request Domains" Restriction Bypass via AI Agents MCP Connector
  advisoryas classification.technologies
• Ruby SSE Session Poisoning
  advisoryas classification.technologies
• Missing authentication on MCP HTTP endpoint allows unauthenticated privileged tool calls
  advisoryas classification.technologies
• DNS TOCTOU race condition causes SSRF protection bypass (`/admin/gateways/test`)
  advisoryas classification.technologies
• Flowise MCP Security Bypass Enables RCE
  advisoryas classification.technologies
• CVE-2026-44895 - GitHub Advisory Database
  advisoryas classification.technologies
• Warning: Vulnerability in Starlette framework and related frameworks like FastAPI exposes millions of servers to authentication bypass, Patch Immediately! | CCB Belgium
  advisoryas classification.technologies
• CVE-2026-87912 and CVE-2026-87913: Missing S3 bucket ownership verification in the AWS Security Agent plugin for aws-agents-for-devsecops and MCP Server
  advisoryas classification.technologies
• CVE-2026-87911 - Read-only enforcement bypass enabling operating system command execution in the SQL validation component of Amazon awslabs postgres-mcp-server
  advisoryas classification.technologies
• No-Box Vulnerability Analysis: Description-only Detection of Indirect Prompt Injection Vulnerabilities in MCP Servers
  paperas classification.technologies
• ShareLock: A Stealthy Multi-Tool Threshold Poisoning Attack Against MCP
  paperas affected productsIt is written here as Model Context Protocol (MCP).

  Open protocol bridging LLMs with external tools; foundational to modern agent ecosystems

• A First Measurement Study on Authentication Security in Real-World Remote MCP Servers
  paperas classification.technologies
• MCPTox: A Benchmark for Tool Poisoning Attack on Real-World MCP Servers
  paperas classification.technologies
• awarexone/Agentic-Bug-Hunter
  repoas classification.technologies
• Solving the Identity Crisis for AI Agents (Uber)
  articleas classification.technologies
• OWASP Agentic Skills Top 10 (AST10)
  repoas classification.technologies
• OWASP Agentic Skills Top 10 (AST10)
  otheras toolsIt is written here as Model Context Protocol (MCP).

  Protocol layer defining how the model talks to tools, distinguished from the agentic skills execution layer that AST10 addresses

• OWASP Agentic Skills Top 10
  paperas classification.technologies
• SANDWORM_MODE: Dissecting a Multi-Stage npm Supply Chain Attack | Blog | Endor Labs
  writeupas classification.technologies
• cyanheads/osv-advisory-mcp-server
  repoas classification.technologies
• Security Hardening Release for Splunk Apps and Add-ons - August 2026
  advisoryas classification.technologies
• Tencent/AI-Infra-Guard
  repoas classification.technologies
• affaan-m/agentshield
  repoas classification.technologies
• h5i-dev/awesome-ai-agent-incidents
  repoas classification.technologies
• CVE-2026-12773
  advisoryas classification.technologies
• Deadbugz: Currently Active MCP Supply-Chain Campaign
  writeupas classification.technologies
• CVE-2026-49291 | THREATINT
  advisoryas classification.technologies
• OWASP MCP Top 10 | OWASP Foundation
  advisoryas affected productsIt is written here as Model Context Protocol (MCP).

  framework for defining operational, contextual, and behavioral boundaries of AI models; MCP-enabled systems including agentic AI, model chaining, multi-modal orchestration, and dynamic role assignment scenarios

• AgentBaiting: How Fake AI Skills Deliver Malware at Scale
  writeupas classification.technologies
• Grafana MCP server-side request forgery via X-Grafana-URL header (grafana_api_request) | Grafana Labs
  advisoryas classification.technologies
• 0x4m4/hexstrike-ai
  repoas classification.technologies
• agent-egress-bench
  repoas classification.technologies
• Dev Machine Guard Now Supports Linux - StepSecurity
  articleas classification.technologies
• Dev Machine Guard Now Supports Windows - StepSecurity
  articleas classification.technologies
• OWASP Agentic Skills Top 10 (AST10)
  otheras toolsIt is written here as Model Context Protocol (MCP).

  Tool layer defining how the model talks to tools; contrasted with AST10 which covers what those tools actually do

In every workspace you can read 500