git:20261005.8a9b127 to git:20261005.4d7d33a

1 added, 1 removed. Audit A to A.

@AGENTS.md

## Claude Code notes

- When you edit this guide, load the bundled `/claude-api` skill before you state a model fact, then confirm the fact on its Tier 1 source from the table in `AGENTS.md`. The guide states no prices or plan limits.

- - This repo's example hook, when `.claude/settings.json` registers it, can run in your session with the user's full permissions: `.claude/hooks/notification.py` speaks a message aloud when Claude Code needs input, using the ElevenLabs or OpenAI API when those keys are in `.env` or the shell environment. Never commit `.env`, and pass `--settings '{"disableAllHooks": true}'` when you script `claude -p` in this clone ([SECURITY.md](SECURITY.md)).

+ - This repo's root `.claude/` holds no hooks and no allow rules: only a deny rule for reading `.env` files, a skill and a workflow ([SECURITY.md](SECURITY.md)). `examples/` is inert until copied; review every file before copying it. Never add a `.claude/`, `CLAUDE.md`, `AGENTS.md` or `.mcp.json` under `examples/`.