Paper builds a 24-case dataset of MCP security vulnerabilities
Authors say 87.5 percent are Critical or High, and injection plus path traversal account for 66.7 percent.
Stories per day · 30 days
4 RSS
Relatedmcp5anthropic2agents1ai-agents1canva1canva-craft1capability-laundering1cloudflare1commentary1credentials1cve-dataset1design-arena1figma1filesystem1image-generation1injection1json-rpc1path-traversal1
15 stories, newest first · one card per story
Stories tagged #model-context-protocol
Ben Hylak jokes every product should be a headless MCP agent
He carves out his own product, saying customers love its interface and the company.
Canva launches Craft image model that plans designs first
Craft is available globally in Dream Lab, outputs up to 2K across more than 30 aspect ratios, and ranks fourth on Design Arena.
1
Figma adds MCP partners as co-founder calls the protocol imperfect
Dylan Field said wider MCP access to Figma data raises abuse risk and favors frontier labs.
WhatsApp MCP server offered free on Cloudflare
The feed only links a GitHub project and shows 23 points and 13 comments, with no technical details.
My Credentials, Please: Detecting Filesystem Capability Laundering in MCP Servers at Scale
Researchers find filesystem capability laundering in MCP servers, including two CVEs in Anthropic's Git server.
Welcome to the Jungle: What We Found Inside 15,465 Public MCP Servers
OX Security's review of 15,465 public MCP servers found no marketplace vetting and easy domain takeover risks.
12
MCP Python SDK OAuth Flaw Lets Malicious Servers Hijack AI Agent Accounts
Malicious MCP servers can hijack OAuth tokens from vulnerable MCP Python SDK clients to take over AI agent accounts; patched in 1.30.0 and 2.2.0.
3
Recorded Future Launches MCP, the Intelligence Layer for Agentic Security Operations
Recorded Future made its MCP generally available so AI agents can query its threat intelligence.
5
Salesforce Agentforce: Bridging the Enterprise AI Gap from ‘Vibe Coding’ to Battle-Tested Orchestration
Salesforce pitches Agentforce as an enterprise agent platform with testing, observability, and deterministic gating; Southwest Airlines reports $6M annual savings and 45% autonomous resolution.
10
Google will now let any AI agent run your smart home
Google's Home MCP integration lets third-party AI agents like Claude control smart home devices and analyze home data.
1
Rubrik MCP gives AI agents controlled access to security intelligence
Rubrik launched MCP support exposing Rubrik Security Cloud APIs to enterprise AI agents with RBAC, configurable permissions, and OWASP MCP Top 10 guardrails.
OpenAI Builds ‘Defense Factory’ Where AI Agents Continuously Find and Fix Vulnerabilities
OpenAI unveils a Defense Factory where AI agents continuously discover, validate, and fix vulnerabilities, integrating GitHub, Snyk, Semgrep, Tenable, and ServiceNow.
3
datasette-mcp 0.2
Simon Willison released datasette-mcp 0.2, changing MCP SQL results to object arrays for better small-model usability.
[remote] mcp-server-kubernetes 3.8.x - Argument Injection
Argument injection in mcp-server-kubernetes 3.8.x MCP server could allow AI clients to execute unintended Kubernetes operations.