karolkozer

Agentwhy — AI skill for Claude Code

AI community

Why your AI agent reached for your secrets - from Claude Code and Codex transcripts.

How to install Agentwhy

This entry records only its repository, not the path inside it, so there is no exact command to give. Open karolkozer/agentwhy and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Agentwhy does

Why your AI agent reached for your secrets - from Claude Code and Codex transcripts.

Alternatives in AI

  • Wiki Synthesize — Synthesize wiki/sources/ .md pages from raw session transcripts using the configured LLM backend 375 ★
  • Engineering Notebook — A Bun CLI that ingests Claude Code and Codex session transcripts, generates LLM-powered daily summaries, and s 320 ★
  • Honeycomb — AI coding agents forget 109 ★

README

agentwhy

Your AI agent reads your files. Find out which ones, and why.

agentwhy reads the conversations Claude Code and Codex already save on your computer and shows which of your private files (.env, keys, secrets) an agent opened, the task that led to it, and which key to change.

npx @agentwhy/cli
One command · runs on your computer · sends nothing anywhere · Apache 2.0

agentwhy: Your AI opens your files. Now you'll know which ones.


What you see

You run one command in your project. A page opens in your browser. Most days it says there is nothing to do. When there is, it tells you what to fix. The pictures below are of a demo project.

**Conversations.** Every time your AI worked for you this week, and which of those times it read something private.

![The Conversations page: nine conversations this week, three of them marked as needing attention](docs/screenshots/conversations.png)

**Your to-do list.** One conversation, opened: the private file your AI read, and a **Fix it** button that walks you through making it safe.

![The to-do list of one conversation: a private .env file was opened, with a Fix it button](docs/screenshots/to-do-list.png)

**What happened.** The story of one file, step by step: who read it, where it was repeated, and where it ended up.

![What happened to this file, as a story: a helper read the keys, repeated them in a message and passed them back](docs/screenshots/what-happened-story.png)

The same story as a diagram, from your request to the file:

![What happened to this file, as a diagram: you, your AI, a helper, the .env file,