Mantis Hits GitHub: 31 Composable Security Skills Form an Agent Vulnerability-Review Pipeline
Saboo_Shubham_ · x · 2026-09-10
The Mantis GitHub repo (1.1k stars) reveals a 31-skill modular architecture — plan, threat-model, reproduce, patch, report, critic, meta-agent, dedupe and more — composing a portable security-review harness for AI coding agents. The authors note it is not an official Google product; it ships with AGENTS.md and a schema for agent integration.
Related event: Google Engineers Open-Source Mantis AI Security Toolkit(2 posts)→
More from coding & agent
- ScreenContext: free open-source Mac recorder that feeds screen videos to AI agents as context — MarcusSchiesser · 2026-09-10
- Dev: Plan 9's 30-year-old 9P protocol fits the external shape of agent systems — remilouf · 2026-09-10
- No frameworks, no magic: turning your work into a library the AI can actually search — dfinke · 2026-09-10
- Personal AGI experiment: a tiny harness, five tools, Markdown, one LLM to give agents memory — dfinke · 2026-09-10
- 16x Microsoft MVP livestreams Personal AGI: five tools and Markdown to make agents compound knowledge — dfinke · 2026-09-10
- asc CLI ships per-release changelogs designed for AI agents to consume — rudrank · 2026-09-10