GitHub and Microsoft's MDASH pilot turns thousands of AI vuln findings into 99 verified fixes
mustafasuleyman · x · 2026-11-18
AI makes finding vulnerabilities fast and cheap, but maintainers are drowning in duplicate and false AI reports. GitHub and Microsoft piloted Codename MDASH, Microsoft's multi-model agentic scanning harness using the MAI Cyber model, with an AI triage agent from GitHub Security Lab filtering findings and humans verifying them. Across six widely used open source projects, thousands of raw AI findings became 99 checked, private reports with suggested fixes—maintainers have already fixed more than half. Advice for anyone pointing a new model at a project: ask maintainers first, and come with fixes, not just findings.
Related event: GitHub and Microsoft Pilot MDASH to Filter AI Bug Reports(4 posts)→
More from coding & agent
- Code Mode turns 1: every major agent harness has adopted it — and kept the name — ritakozlov · 2026-11-18
- Simon Willison open-sources three external sandboxes for safer coding agents — Simon Willison · 2026-11-18
- OpenAI updates API dashboard with per-model usage, rate-limit and error tracking — OpenAIDevs · 2026-11-18
- 'Proof-Carrying Calls' Idea: Make Agents Certify They Don't Break Your Invariants — tianyin_xu · 2026-11-18
- Devin Platform Adds a 'Poke' Bot Template: An Engineering Agent with Personality — marvinvonhagen · 2026-11-18
- AWS cuts agent token costs 71% and latency 80% on 23MB JSON workloads with three strategies — AWS ML Blog · 2026-11-18