add_hooklisted

> **Claude Code specific**: this skill manages Claude Code configuration > (settings.json wiring). Other agents do not consume what it produces. ## Step 0 — Confirm the current schema (best-effort) The event names and output schema below are a snapshot and can lag Claude Code releases. **If** a docs-lookup tool is available to you — the Context7 MCP, or WebFetch when it's permitted — confirm the event name / output field against the current reference (<https://docs.agents.com/en/docs/claude-code/hooks>) before relying on it. This skill's `allowed-tools` does not grant those tools, so skip this step cleanly whenever the tool is unavailable, unapproved, egress is disabled (`--no-egress` / air-gapped), or the lookup fails — fall back to the embedded reference below. Don't request extra permissions and never block on it. ## Step 1 — Choose an event Pick the event that matches when the hook should fire: **Tool execution** (most common): - `PreToolUse` — before a tool runs; output block JSON to block it - `PostToolUse` — after a tool runs; cannot block, can log or validate - `PostToolBatch` — after a batch of parallel tool calls completes - `PostToolUseFailure` — after a tool fails - `PermissionRequest` — when Claude asks for permission; can auto-approve - `PermissionDenied` — when permission is denied; can request a retry **Session lifecycle:** - `SessionStart` — when a session begins - `SessionEnd` — when a session closes - `PreCompact` — before context compaction (can blo