chore: bootstrap repo with chunks 1 and 2

Establishes the global AI development config repo from scratch:

- Chunk 1: repo skeleton, install.sh, statusline, deploy manifest
- Chunk 2: core instructions (coding/git/testing), CLAUDE.md rewrite
  (always-on + content index two-tier model), docs restructure,
  6 ADRs, ROADMAP.md, .gitkeep placeholders
- Bootstrap skills in .claude/skills/ (to be catalogued and migrated
  to .agents/skills/ in Chunk 3)

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

tests/test-statusline.sh

@@ -0,0 +1,72 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+REPO_ROOT="$(cd "$(dirname "${BASH_SOURCE[0]}")/.." && pwd)"
+SCRIPT="$REPO_ROOT/providers/claude-code/statusline-command.sh"
+PASS=0
+FAIL=0
+
+pass() { echo "  PASS: $1"; PASS=$((PASS + 1)); }
+fail() { echo "  FAIL: $1"; FAIL=$((FAIL + 1)); }
+
+# Strip ANSI escape codes so assertions work on plain text
+strip_ansi() { sed 's/\x1b\[[0-9;]*m//g'; }
+
+# Pipe JSON to the script and return stripped output
+run() { echo "$1" | bash "$SCRIPT" | strip_ansi; }
+
+# Baseline input covering all segments
+FULL='{"workspace":{"current_dir":"/home/user/myproject"},"model":{"display_name":"Sonnet 4.6"},"context_window":{"used_percentage":20,"total_input_tokens":15000,"total_output_tokens":5000},"cost":{"total_cost_usd":0.25}}'
+
+echo "--- assembly ---"
+
+out=$(run "$FULL")
+
+echo "$out" | grep -q "myproject"                  && pass "shows basename of working directory"      || fail "dir missing — got: $out"
+echo "$out" | grep -q " · "                        && pass "segments joined with ' · '"               || fail "separator missing — got: $out"
+echo "$out" | grep -q "Sonnet 4.6"                 && pass "shows model name"                         || fail "model name missing — got: $out"
+echo "$out" | grep -q "context 20%"                && pass "shows context % with 'context' label"     || fail "context label wrong — got: $out"
+
+echo ""
+echo "--- token formatting ---"
+
+# 15k + 5k = 20k → shown as "20k tokens"
+echo "$out" | grep -q "20k tokens"                 && pass "formats total tokens as Xk when >= 1000"  || fail "Xk format wrong — got: $out"
+
+# Sub-1000 total: 500 + 300 = 800 → shown as "800 tokens"
+LOW='{"workspace":{"current_dir":"/x"},"context_window":{"total_input_tokens":500,"total_output_tokens":300}}'
+out_low=$(run "$LOW")
+echo "$out_low" | grep -q "800 tokens"             && pass "shows raw count when total tokens < 1000" || fail "sub-1000 format wrong — got: $out_low"
+
+echo ""
+echo "--- cost formatting ---"
+
+# 0.25 → 25¢
+echo "$out" | grep -q "25¢"                        && pass "formats sub-dollar cost as cents (¢)"     || fail "cents format wrong — got: $out"
+
+# 1.71 → $1.71
+DOLLAR='{"workspace":{"current_dir":"/x"},"cost":{"total_cost_usd":1.71},"context_window":{"total_input_tokens":0,"total_output_tokens":0}}'
+out_dollar=$(run "$DOLLAR")
+echo "$out_dollar" | grep -qF '$1.71'              && pass "formats dollar-plus cost as \$X.XX"       || fail "dollar format wrong — got: $out_dollar"
+
+echo ""
+echo "--- missing fields omitted ---"
+
+# No cost field → no ¢ or $ in output
+NO_COST='{"workspace":{"current_dir":"/x"},"context_window":{"total_input_tokens":0,"total_output_tokens":0}}'
+out_nocost=$(run "$NO_COST")
+! echo "$out_nocost" | grep -qE '[¢$]'             && pass "omits cost segment when cost absent"      || fail "cost shown unexpectedly — got: $out_nocost"
+
+# Zero tokens → no token segment
+ZERO_TOK='{"workspace":{"current_dir":"/x"},"context_window":{"total_input_tokens":0,"total_output_tokens":0}}'
+out_zerotok=$(run "$ZERO_TOK")
+! echo "$out_zerotok" | grep -q "tokens"           && pass "omits token segment when total is zero"   || fail "token segment shown unexpectedly — got: $out_zerotok"
+
+echo ""
+echo "--- executable bit ---"
+
+[[ -x "$SCRIPT" ]]                                 && pass "statusline-command.sh is executable"      || fail "statusline-command.sh is not executable"
+
+echo ""
+echo "Results: $PASS passed, $FAIL failed"
+[[ $FAIL -eq 0 ]]