Can an LLM write EasyCrypt proofs, and this repo ships the MCP workflow, benchmarks, and replay logs to find out.
trending8
01 02 Research prototype for EasyCrypt lemma proofs, using an MCP agent to spit out admit-free scripts and replay-check them.03 Unicode TAG-Block Concealment of Tool-Metadata Payloads in the Model Context Protocol: An Approval-View Fidelity Gap Across Three Independent Server Implementations arxiv.orgCan MCP tool approvals hide Unicode TAG payloads from humans while still feeding them to the model? Apparently yes.04 [tl;dr sec] #335 - Prompt Injection as Role Confusion, PHP Ecosystem Security, New MCP Spec tldrsec.comPrompt injection, apparently, is role confusion, in a newsletter that also hits PHP hardening and the new MCP spec.05 Red Hat Delivers Post-Quantum Readiness and AI-Powered Automation with Latest Versions of Red Hat Enterprise Linux redhat.comRed Hat's RHEL 10.2/9.8 pitch adds post-quantum crypto, confidential computing, and AI-assisted admin to the usual distro churn.06 From Tool Connection to Execution Control: Benchmarking Security Invariants in MCP-Style Agent Runtimes arxiv.orgEight security invariants, because MCP tool connections apparently weren’t enough, and the paper benchmarks HCP against baselines.07 A Microsoft IFC writeup for agents, with label propagation, policy checks, and prototypes for MCP, Copilot CLI, and Agent Framework.08 The 2026 Guide to Post-Quantum AI Infrastructure Security: Securing MCP Deployments securityboulevard.comGuide to hardening MCP deployments with mTLS, sandboxing, policy controls, and hybrid PQ crypto for the usual future-proofing theater.