mantis: a set by google

mantis-threat-model official google skill Turns the Mantis knowledge base into a project threat model: trust boundaries, entry points, attacker profiles. Useful once the architecture is written down but the threat model is not. Claude Code CX Codex Cursor +1 1,800installs mantis-plan official google skill Generates a targeted security review plan based on your threat model and historical findings. Run at the start of a review campaign to map code boundaries and prioritize what to check. Claude Code CX Codex Cursor +1 1,600installs mantis-summarize official google skill Builds a security-focused codebase map with mantis-summary.md files in each directory. Reduces token overhead during planning and research. Claude Code CX Codex Cursor +1 1,600installs mantis-meta-agent official google skill A campaign supervisor for Mantis security review: it keeps the loop running pass after pass and holds state and snapshots. Built for reviews that run long unattended. Claude Code CX Codex Cursor +1 1,500installs mantis-reflect official google skill Extracts insights, failures, and false assumptions from Mantis agent transcripts, appending them to workspace/learnings.jsonl. Claude Code CX Codex Cursor +1 1,500installs mantis-pipeline-adapter official google skill A consulting skill that walks you through building your own deterministic harness for running Mantis skills. Claude Code CX Codex Cursor +1 1,500installs mantis-patch official google skill Generates minimal security fixes in isolated environments, applies patches, and verifies them against reproduced vulnerability findings. Claude Code CX Codex Cursor +1 1,500installs mantis-calibrate official google skill Turns confirmed findings into a final risk score from 1 to 10, weighing reproduction evidence against architectural impact. Runs at the end of the pipeline, before the report. Claude Code CX Codex Cursor +1 1,500installs mantis-reproduce official google skill Generates and executes crash reproducers in isolated sandboxes to verify security flaws. It helps confirm findings before patching. Claude Code CX Codex Cursor +1 1,500installs mantis-dedupe official google skill Consolidates duplicate security findings before review. Use it after raw findings exist and several reports may describe the same flaw. Claude Code CX Codex Cursor +1 1,500installs mantis-report official google skill Transforms verified security findings and exploit chains into a professional markdown report. Use at the end of your review cycle to generate stakeholder-facing documentation. Claude Code CX Codex Cursor +1 1,500installs mantis-researcher official google skill Audits production source code against workspace/plan.json. It inspects target files for boundary checks, missing sanitization, and interface violations without modifying the codebase. Claude Code CX Codex Cursor +1 1,500installs