TheMorpheus407/RepoLens

Multi-lens code audit tool — 280 expert AI agents for code review, security testing, and infrastructure auditing

What it solves

RepoLens automates complex, multi-domain code auditing by running specialized "lenses" against repositories, live servers, Android APKs, or product specifications. It identifies security vulnerabilities, backlog tasks, and code quality improvements, helping developers manage large-scale technical debt and security posture without manual inspection.

How it works

RepoLens acts as an orchestration engine that coordinates various AI agent backends (such as Claude Code, OpenAI Codex, Cursor, or Google Antigravity) to perform deep analysis. It executes hundreds of specialized audit lenses across dozens of domains, managing the lifecycle of each agent request, handling timeouts, and optionally creating issues directly on platforms like GitHub, GitLab, Gitea, or Forgejo. It supports both parallel execution for speed and local-only modes for security-sensitive environments.

Who it’s for

This tool is designed for developers, security engineers, and technical leads who need to perform comprehensive, automated audits on large or complex codebases and require deep, agent-driven insights rather than simple static analysis.

Highlights

  • Runs 338 specialized audit lenses across 34 distinct domains.
  • Integrates with multiple agent backends including Claude, Cursor, OpenAI Codex, and Google Antigravity.
  • Supports remote forge integration to automatically file issues on GitHub, GitLab, Gitea, and Forgejo.
  • Includes cost-management features like --max-cost and --flat-rate to help users monitor and control API usage.
  • Provides a --dry-run mode to preview audit scope and estimated wall-clock time before execution.

Related

  • Project
    NewT123-WM/tnlearn

    Tnlearn is a Python library that uses symbolic regression to discover task-specific neuron designs, allowing users to build custom, high-performance neural networks inspired by biological neuronal diversity.

  • Project
    ChristopherKahler/base

    A local knowledge graph engine that gives AI coding agents persistent, context-aware memory by automatically injecting relevant project rules, code structure, and decision history into their workflow.

  • Project
    cacheplane/b4run

    B4.run is a TypeScript meta-framework for building LLM agents on top of LangGraph.js, adding file-system routes, generated types, local test loops, persistence primitives, and build targets. It lets developers author application shape instead of wiring route plumbing, with options for tool approval, permissions, and sandboxing.

  • Project
    remorses/zele

    A command-line interface for managing email and calendars that provides structured YAML output, making it ideal for automating tasks and integrating with AI agents.