Research and Blog

- Product
Manifold Launches Cost and Usage Controls: Your Agents Now Come With a Budget
Oleksandr Yaremchuk
Read the article All articles

Research
Did the AI Really Escape Its Sandbox?
Markus GierlingerRead
Research
A Cautionary Tale on Git Path Resolution: From Startup to Arbitrary Command Execution in Mistral Vibe (CVE-2026-122133)
Francisco RosalesRead
Product
Remove Risky Skills, Plugins and MCP Servers From Every Endpoint
Oleksandr YaremchukRead
Research
Who Clicks 'Yes'? What 6,000+ Agent Add-Ons Say About Approval
Ax SharmaRead
Research
How Long Do AI Agent Incidents Take to Come to Light?
Bastien EymeryRead
Research
What Counts as an AI Agent Incident?
Bastien EymeryRead
Research
What 66 AI Agent Incidents Tell Us
Bastien EymeryRead
Research
What is your agent configured to run?
Francisco RosalesRead
Product
Security in the Loop, Not in the Way: Guiding What AI Agents Do
Oleksandr YaremchukRead
Product
Introducing the Agent Incident Ledger: A Rogues' Gallery of AI Gone Wrong
Bastien Eymery and Neal SwaelensRead
Research
AI agents have few qualms with log tampering
Cody NashRead
Research
From Model Risk to Action Risk
Oleksandr YaremchukRead
Product
Manifold now rates the editor extensions your developers install
Oleksandr YaremchukRead
Research
Personal agents live in iMessage and WhatsApp. Your EDR doesn’t.
Ax SharmaRead
Research
One vendor, three platforms: governing Claude Code and Cowork
Guido BonomiRead
Research
The High-Risk Domains of Agent Skills and Where They Lead
Swapnil Patil and Cody NashRead
Product
Manifold Enables Security Teams to See Which AI Agent Threats to Act On First
Oleksandr YaremchukRead
Research
Four Questions to Ask Before Calling an AI Agent Rogue
Ax SharmaRead
Research
Ask a Model If Code Is Malicious and It Reaches for Its Morals
Cody NashRead
Research
Your Review Approved Version 1.0. Your Agents May Be Running Whatever Shipped This Morning.
Oleksandr YaremchukRead
Research
Jev, System One Models, and How to Use Them in Security
Cody Nash and Markus GierlingerRead
Research
Rogue Agents Targeted a Second Australian Health Dashboard, urlscan.io Index Shows
Ax SharmaRead
Research
Over 350,000 GitHub Files Cite Placeholder Domains Serving Scams
Cody NashRead
Product
Blocking Is the Only Answer Most Security Tools Have. "Ask" Lets the Developer Decide.
Neal SwaelensRead![The third-party[.]com Domain Is Serving a ClickFix Lure to Windows Users](/_next/image?url=https%3A%2F%2Fcdn.sanity.io%2Fimages%2Fphwxcc13%2Fproduction%2Fe4d506fb5667026cd9a7c5a9a8536cc6843b7d56-1280x858.webp%3Frect%3D0%2C3%2C1280%2C853%26w%3D800%26h%3D533%26auto%3Dformat&w=3840&q=75)
Research
The third-party[.]com Domain Is Serving a ClickFix Lure to Windows Users
Ax SharmaRead
News
AI Agents Moved Into the Browser, Make Sure Your Governance Follows
Neal SwaelensRead
News
Joe Sullivan Joins Manifold's Board
Neal SwaelensRead
Research
GitSpawn: A Single Flaw Lets Untrusted Repos Run Code in Claude Code, Codex, Cursor, and Grok
Francisco RosalesRead
Research
OpenAI & Hugging Face: Why the Chain-of-Thought Police Won't Save You
Cody NashRead
Research
What to Think About curl | bash Now That AI Agents Run It.
Ax Sharma and Cody NashRead
Research
Cursor CLI Ran Untrusted Repository Code With the Sandbox Switched Off
Francisco RosalesRead
Research
When a Conversation Wanders, Do an Agent's Guardrails Slip?
Cody NashRead
Research
77 "Evil Twin" Open VSX Extensions: 19 Copy Private Repo and CI Data to a New Domain
Ax Sharma and Cody NashRead
Research
OpenAI and Anthropic Both Disclosed the Same Thing: The Agent Is the Threat Actor Now
Nate DemuthRead
Research
Anyone With an Account Could Run Commands on PewDiePie's AI Workspace (CVSS 9.9)
Francisco RosalesRead
Research
Lethal Trifecta Noise: Why It Fits Every Agent You Deploy
Nate DemuthRead
Research
The "Restricted" Deployment That Wasn't: Two Access-Control Bugs in Community-Built mcp-atlassian
Francisco RosalesRead
Research
A Beaconing Counterfeit Extension on the VS Code Marketplace: The 'Markdown All Pro' Extension
Ax Sharma and Cody NashRead
Research
When Your AI Reviewer Works for the Attacker: A Confused-Deputy Bug in Microsoft's Azure DevOps MCP Server
Francisco RosalesRead
Research
Intent Drift Noise: Why Session-Level Intent Drift Falls Apart in Production
Nate DemuthRead
Research
ClaudeBleed Reopened: Browser Extensions Can Still Push Claude for Chrome to Read Your Gmail
Ax SharmaRead
Research
Scope Squatting: Those @openclaw and @clawhub Plugins Aren't Officially Theirs
Ax SharmaRead
Research
CVE-2026-52024, CVE-2026-52025: We Bypassed Two of Cline's Safety Checks to Run Code (4.2M Installs)
Ax SharmaRead
Research
Is Prompt Injection a Vulnerability?
Ax SharmaRead
Research
10,000 Vulnerabilities in 30 Days. Picture This Threat Inside Your Stack.
Oleksandr YaremchukRead
Research
CVE-2026-54052: A Wrong Number in n8n-mcp Leaked Your Neighbor's Credentials (CVSS 9.6)
Ax Sharma and Francisco RosalesRead
Research
What the AI Village Disasters Reveal About Runtime Security
Neal SwaelensRead
Research
When "Read-Only Mode" Isn't: CVE-2026-46519 in mcp-server-kubernetes
Ax Sharma and Francisco RosalesRead
Research
One AI Security Scanner Flagged 40% of Skills as Malicious. Most Were Fine.
Ax SharmaRead
Product
Manifold Adds MCP Servers to Manifest Supply Chain Intelligence
Oleksandr Yaremchuk and Neal SwaelensRead
Research
30 ClawHub Skills Are Quietly Recruiting Your AI Agent Into a Crypto Swarm
Ax SharmaRead
Research
An OpenClaw Skill With an SVG Can Steal Your Cookies
Ax SharmaRead
Research
The Gateway Gap: Why AI Agent Security Needs More Than a Chokepoint
Oleksandr Yaremchuk and Nate DemuthRead
Research
Two Git Commands Fooled Claude Into Merging Malicious Code
Ax Sharma and Oleksandr YaremchukRead
Product
Introducing Manifest: Supply Chain Intelligence for the AI Agent Ecosystem
Oleksandr Yaremchuk and Neal SwaelensRead
Research
Non-Adversarial Agent Harm: The Overlooked Insider Threat
Neal SwaelensRead
Research
Why Your EDR Has No Clue What Your Agents Are Up To
Nate Demuth and Neal SwaelensRead
Research
Why AI Guardrails Can’t Secure Your Agents: The Classification Fallacy
Neal Swaelens and Oleksandr YaremchukRead
News
Manifold Raises $8M to Secure AI Agents on Endpoints
Manifold TeamRead
Research
The Sandbox Illusion: How Do You Isolate Agents That Think Outside the Box?
Neal Swaelens and Oleksandr YaremchukRead










