feat(claude): add 6 lifecycle hooks reinforcing template hard rules

Project-level Claude Code hooks committed to .claude/settings.json with
scripts under .claude/hooks/. Three tiers:

Tier 1 — hard guards (exit 2 to block the tool call):
- bash-guard.sh: blocks bypass flags (verify-skip, sign-skip), forceful
  push variants, destructive resets, force clean, working-tree-wipe
  checkouts/restores, force branch delete, amend, and rm -rf against
  root or home. Reinforces CLAUDE.md Git Safety Protocol.
- generator-first-nudge.sh: blocks creating a new top-level
  packages/<name> or apps/<name> directory by hand. Allows working
  inside an existing package. Reinforces the non-negotiable
  generator-first rule.

Tier 2 — context injection (stdout becomes additional context):
- session-start.sh: prints glossary, AGENTS.md, workflow CLI, and
  conformance pointers on session boot.
- prompt-context.sh: keyword-matches the user prompt against eight
  concept groups (events, realtime, audit, instrumentation, manifest,
  workflow, DI, boundaries) and injects the relevant ADR + rule
  pointers for the turn.

Tier 3 — side-effect automation:
- post-manifest-edit.sh: when Edit/Write touches feature.manifest.ts,
  prints the manifest-first ordering reminder plus the per-feature
  verify commands.
- stop-check-manifest-tests.sh: at agent Stop time, if the working
  tree has manifest changes but no sibling test changes, exits 2 to
  force continuation. Loop-guarded via stop_hook_active.

All hooks are bash + jq, use CLAUDE_PROJECT_DIR for safety, and were
smoke-tested end-to-end (block + allow paths both verified).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

.claude/hooks/bash-guard.sh

@@ -0,0 +1,44 @@
+#!/usr/bin/env bash
+# Tier 1 — blocks dangerous shell invocations the agent shouldn't run
+# autonomously. Reads PreToolUse JSON on stdin; exits 2 with stderr to block,
+# 0 to allow. Reinforces the Git Safety Protocol in CLAUDE.md.
+
+set -euo pipefail
+
+input=$(cat)
+cmd=$(printf '%s' "$input" | jq -r '.tool_input.command // ""')
+
+blocks=(
+  '(^|[[:space:]])--no-verify([[:space:]]|$)'
+  '(^|[[:space:]])--no-gpg-sign([[:space:]]|$)'
+  'git[[:space:]]+push[[:space:]]+([^&|;]*[[:space:]])?(-f|--force)([[:space:]]|$)'
+  'git[[:space:]]+reset[[:space:]]+[^&|;]*--hard'
+  'git[[:space:]]+clean[[:space:]]+-[a-zA-Z]*f'
+  'git[[:space:]]+checkout[[:space:]]+\.([[:space:]]|$)'
+  'git[[:space:]]+restore[[:space:]]+\.([[:space:]]|$)'
+  'git[[:space:]]+branch[[:space:]]+-D'
+  'git[[:space:]]+commit[[:space:]]+[^&|;]*--amend'
+  'rm[[:space:]]+-rf?[[:space:]]+/'
+  'rm[[:space:]]+-rf?[[:space:]]+~'
+  'rm[[:space:]]+-rf?[[:space:]]+\$HOME'
+)
+
+for pattern in "${blocks[@]}"; do
+  if [[ "$cmd" =~ $pattern ]]; then
+    cat >&2 <<EOF
+BLOCKED by .claude/hooks/bash-guard.sh
+
+This template forbids the agent from running this autonomously:
+  Pattern: ${pattern}
+  Command: ${cmd}
+
+If the user has explicitly authorized this action this turn, ask them to
+run it themselves (\`! <command>\` in the prompt) or document the override
+in their request. See CLAUDE.md → "Executing actions with care" and the
+Git Safety Protocol section.
+EOF
+    exit 2
+  fi
+done
+
+exit 0