Tag
开源了 PuedoBot，一个基于确定性扫描加双 pass LLM 流水线的 CLI 安全审计工具，第二个 pass 专门反驳第一个 pass 的发现以减少 SAST 误报。
Neal Bridges released VulnHunter, an open-source fork of Capital One's agentic AI vulnerability-hunting and PoC-development kit, updated to run on any agent harness with sandboxed containerized exploit validation and measured-impact PoCs.
A developer compares GLM 5.2 with Fable 5 and Opus 4.8 on OpenRouter, finding GLM 5.2 cost-effective for code audits and reviews but not for long unattended builds due to accumulating test-error-edit costs.
shadcn released the Agent Skill project improve, which has high-cost models perform code auditing and planning while low-cost models execute, forming an installable, orchestrated system with an execution closed loop.
Introduces an 8-stage vulnerability discovery agent based on Claude subscription, reproducing the core ideas of the Cloudflare Project Glasswing paper. It uses multiple narrow agents, intentional disagreement verification, and reachability gating to break down exploit analysis into precise steps, without the need for API keys.
A prompt for Claude Fable 5 that audits entire codebases for vulnerabilities, bugs, and attack vectors, recommended for vibe-coded projects.
Introducing /improve, a tool that uses Claude's most capable model to audit your codebase, identify bugs, performance issues, tech debt, missing tests, and write plans that cheaper models can execute.
A critique of large Rails codebases in the AI agent era, proposing a shift to skill-based development with agents, markdown skills, and TypeScript for deterministic I/O.
Introducing pi-dynamic-workflows, a tool that enables LLMs to dynamically orchestrate multiple sub-agents for parallel task execution by generating JavaScript scripts, suitable for code audits, large-scale refactoring, and similar scenarios.