CrossBeam Permits — AI-assisted building-permit plan review for cities and builders.

Ask Codex, Gemini, Grok, and 400+ OpenRouter models (Qwen, Kimi, DeepSeek) for second opinions or arbiter-mediated consensus. One MCP server for Claude Code, Codex, Cursor, Kiro, OpenCode. Measures which models earn their seat.

Claude Code plugin for validating implementation plans with parallel review subagents — checks codebase alignment, best practices, feasibility, and blind spots before execution

Browser-first markdown plan review. Start local web review in one command, then let your agent continue through MCP.

[ARCHIVED → github.com/Redloft/redkit] plan-panel + finalize консолидированы в монорепо redkit

Local browser review app for Codex plans

Catch the bad plan before you spend the tokens. Cross-LLM peer review via MCP.

Garry Tan's Mega Plan Review adapted for startup business plans, launch bibles, and fundraising strategies. Same rigor, same three modes (EXPANSION / HOLD SCOPE / REDUCTION), different domain. Stress-test your startup plan like Garry stress-tests his engineering plans.

日常简单 Grill；方案先预审核，只在 ADR 时集中请求决策

Auditable self-running critique loops for AI agents and plan review

Review construction plan sets like a plan checker + owner's rep — permit-set review and active-project document control (revisions, ASIs, RFIs, shop drawings, inspections). Installable Claude skill/plugin.

Source-aware plan-review skill for Claude Code, Codex, and AI coding agents.

A Claude Code skill for iterative, independent plan review before execution. Domain-neutral orchestration with pluggable reviewer prompts (code / judicial PT-BR / generic) and a deterministic Python validator as a hard gate against open findings.

Structured multi-agent review for plans, specs, and code with iterative debate, patch-based auto-apply, and a full audit trail.

3-round cross-agent spec/plan review pipeline. Reviewer audits findings and author settles them across fresh sessions, with a strict final round.

Local multi-CLI plan reviews for AI coding agents.

Adversarial second-opinion reviews for AI coding agents. Auto-gates risky plans before ExitPlanMode and routes them through Codex/Gemini/Opencode for a fresh-lineage critique — with a self-review fallback. Ships as a Claude Code plugin and a portable skill.

VS Code extension that intercepts Plannotator browser opens and displays plan reviews inside VS Code tabs instead of an external browser.

Best Free AI Code Review Tool 2026 – Catch Flawed Plans Before Deployment

An adversarial review system for LLM agents. Shadow personas are observational agents that monitor a primary agent's output and produce independent critique — without participating in the conversation.