Wy

Wy is a command line tool that links AI agent code changes to the specific chat transcript reasons.

Wy is a command line tool designed to help developers track why AI agents made specific code modifications. It reads saved local Codex and Claude Code logs whose working directory is the repository, matching each Git diff hunk to the exact edit record and conversation turn that produced it. Users can pick a file to view changes in order alongside the model name and the reasoning provided by the agent. The tool operates completely offline for its core matching features, while optional fuller explanations use the user's signed-in CLI. It supports developers working with AI coding assistants who need to audit modifications without scrolling through long transcripts.

Wy at a glance

Best for: Software engineer
Runs on: CLI, macOS, Windows, Linux
Pricing: Free and open source under MIT license.
Released: October 2026
Category: AI coding tool
Website: github.com/grandimam/wy

Key features

Reads saved Codex and Claude Code logs
Matches Git diff hunks to agent edits
Displays agent messages above code changes
Shows model names for each modification
Expands to full conversation turns on demand
Generates fuller explanations with cited sources
Operates offline for core matching functions

Use it for

Auditing code changes made by AI agents
Reviewing reasons behind specific file edits
Reading full conversation turns for code modifications

Alternatives to Wy

jevgrep — Reduces coding agent costs by providing targeted, relevant repository context.
Coucou — It provides open-source, real-time agent monitoring and approvals directly from the screen notch.
pstack — Improves AI coding agent outcomes with rigorous workflows.
CodeGraph — It eliminates file crawling, reducing AI agent tool calls by 88% and cutting costs.
Entelligence AI — Inspect agent sessions locally without digging through logs or sharing data.
Context7 — Provides up-to-date documentation for AI agents and code editors.
Crit — It keeps human feedback localized and precise for AI agent code iterations.
ZCode — Integrates GLM-5.3 with multi-agent workflows and remote chat control.
Magpie — It enables AI agents to use models not natively supported by their internal protocols.
Lody — Run coding agents in parallel with isolated worktrees and live diff reviews.
Aegis — It ensures AI coding agents work against verified baselines and proof.
Agent Archive — Hand off coding sessions between Claude Code, Codex, and Cursor across machines.

All Wy alternatives