sandeepsirodia

Neuralyzer — AI skill for Claude Code

AI community

Your AI agent saw your API keys, and saved them in plaintext forever.

How to install Neuralyzer

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

What Neuralyzer does

Your AI agent saw your API keys, and saved them in plaintext forever. Scan, redact and rotate secrets in Claude Code & Codex logs. 0 deps.

Alternatives in AI

  • Vibeproxy — Native macOS menu bar app to use your Claude Code & ChatGPT subscriptions with AI coding tools - no API keys n 3.3k ★
  • Dsh Plugin Subscriptions — Use ChatGPT (Codex), Claude, and Grok (X Premium) subscriptions as DeepSeek Harness LLM providers — OAuth logi 283 ★
  • Kru — Local-first MCP password and credential manager for AI agents 125 ★

README

neuralyzer

Your AI agent saw your API keys. Now it never forgets them. Let's fix that.

CI Zero dependencies Works with Claude Code and Codex MIT

13 kinds of secret · 1 GB scanned in ~23 s · 100% local · 0 dependencies

Remember that time your agent ran `cat .env` to "check the config"?

It worked. You moved on. But every byte of that output was written, verbatim, to a file on your disk:

~/.claude/projects/-Users-you-api/3f2b9c….jsonl

It's still there. So is the `env` dump from last Tuesday. So is the database URL with the password in it. Your agent keeps a perfect diary, and your keys are in it, in plaintext, synced to every backup you own.

**neuralyzer is the flashy thing from *Men in Black*.** Point it at the diary. 📸 The agent forgets.

Before / after

**Before**

$ neuralyzer scan
…/3f2b9c.jsonl:212  anthropic  sk-ant…9f2c
…/3f2b9c.jsonl:212  db-url     Qx81zP…u2Lk
…/rollout-0901.jsonl:88  github  ghp_Za…41Rn

3 secret(s) found.

**After**

$ neuralyzer scrub --apply
redacted 2 secret(s) in …/3f2b9c.jsonl
redacted 1 secret(s) in …/rollout-0901.jsonl

$ neuralyzer scan
0 secret(s) found.

Your transcript still works, and `claude --resume` still works. The key is just `[REDACTED:anthropic]` now.

Try it (10 seconds, read-only)

uvx --from

Related Skills

Agent Secrets — AI skill for Claude Code

Agent Secrets

Let an AI agent use your API keys and passwords without the plaintext ever entering its transcript. A plugin f

AI community
Llmscrub — AI skill for Claude Code

Llmscrub

Sweep LLM agent logs (Claude Code, Codex) for leaked secrets and redact them in place.

AI community
Claude Unlimited — AI skill for Claude Code

Claude Unlimited

Rotate Claude and GPT subscriptions and API keys seamlessly inside the Claude Code CLI — switch on usage limit

AI community
Hush — AI skill for Claude Code

Hush

a secret store for ai agents with one rule: the agent never sees the plaintext. get a secret once into the os

AI community
Phantom Secrets — AI skill for Claude Code

Phantom Secrets

Stop AI coding agents from leaking your API keys. Local proxy + MCP that swaps real secrets for phm_ tokens —

AI community
1password Skill — AI skill for Claude Code

1password Skill

A Claude Code skill for securing API keys with 1Password CLI. Stop hardcoding secrets where your AI tools can

AI community

Related Agents

Secret Purist — Security agent for Claude Code

Secret Purist

The paranoid sentinel of credential security. Use this agent to scan codebases and git history for leaked secr

Security community
Secret Guard — Development agent for Claude Code

Secret Guard

MUST be used whenever the user asks to scan for secrets, API keys, credentials, or tokens before a commit or i

Development community
Secret Hunter — Development agent for Claude Code

Secret Hunter

Hunts publicly-leaked credentials (API tokens, cloud keys, DB strings) that belong to a previously-ingested pr

Development community