@inbrace-tech/tokenline 1.2.1 → 1.2.3

package/README.md

@@ -12,18 +12,17 @@
 
 Run the installer for your AI coding CLI:
 
-**Claude Code**
+**Claude Code Per-project installation (`./.claude/settings.json`)**
 ```bash
-# Per-project installation (./.claude/settings.json)
 npx @inbrace-tech/tokenline init
-
-# Global installation (~/.claude/settings.json)
+```
+**Claude Code Global installation (`~/.claude/settings.json`)**
+```bash
 npx @inbrace-tech/tokenline init --global
 ```
 
-**Antigravity CLI**
+**Antigravity CLI Global installation (`~/.gemini/antigravity-cli/settings.json`)**
 ```bash
-# Global installation (~/.gemini/antigravity-cli/settings.json)
 npx @inbrace-tech/tokenline init --antigravity
 ```
 
@@ -68,7 +67,7 @@ LLMs are stateless — by default, they must reread your entire codebase context
 
 **Prompt Caching** solves this by saving that processed state in the provider's memory:
 - **Cache Write (Cold):** The model reads everything and stores the state. This costs slightly more than base tokens.
-- **Cache Hit (Warm):** If your prompt prefix matches the cached state exactly, the model skips the reading phase. This is **~90% cheaper** and starts responding almost instantly.
+- **Cache Hit (Warm):** If your prompt prefix matches the cached state exactly, the model skips the reading phase. This is **~90% cheaper** (**97.5%** on Claude Fable 5.1 / Mythos 5.1, which bill cache hits at 0.025x base input) and starts responding almost instantly. `tokenline` picks the right multiplier from the active model.
 - **TTL (Time-To-Live):** The cache operates as a **sliding window**. Every cache hit resets the countdown to its full duration for free. The secret to minimizing rate-limit drain is adjusting your workflow pace to keep the cache continuously **HOT**.
 
 ### 💡 Pro-tip: Forcing a 5m TTL to save rate limits

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@inbrace-tech/tokenline",
-  "version": "1.2.1",
+  "version": "1.2.3",
   "description": "A cache-aware statusline for AI coding CLIs — installs tokenline.sh and wires it into your Claude Code settings.",
   "bin": {
     "tokenline": "dist/cli.js"
@@ -37,29 +37,30 @@
     "access": "public"
   },
   "devDependencies": {
-    "@changesets/cli": "^2.31.0",
+    "@changesets/cli": "^3.0.2",
     "@eslint/js": "10.0.1",
-    "@types/node": "24.13.3",
-    "@vitest/eslint-plugin": "1.6.23",
-    "eslint": "10.7.0",
+    "@types/node": "24.13.4",
+    "@vitest/eslint-plugin": "1.6.27",
+    "eslint": "10.10.0",
     "eslint-config-prettier": "10.1.8",
-    "eslint-plugin-n": "18.2.2",
+    "eslint-plugin-n": "18.3.0",
     "eslint-plugin-prettier": "5.5.6",
     "eslint-plugin-simple-import-sort": "13.0.0",
-    "globals": "17.7.0",
-    "prettier": "3.9.5",
+    "globals": "17.12.0",
+    "prettier": "3.9.6",
     "tsup": "^8.5.1",
     "typescript": "5.9.3",
-    "typescript-eslint": "8.64.0",
-    "vitest": "4.1.10"
+    "typescript-eslint": "8.70.0",
+    "vitest": "4.1.11"
   },
   "scripts": {
     "build": "tsup",
-    "typecheck": "tsc --noEmit -p tsconfig.json",
+    "typecheck": "tsc --noEmit -p tsconfig.json && tsc --noEmit -p scripts/tsconfig.json",
     "lint": "eslint . --cache --cache-location node_modules/.cache/.eslintcache --fix",
     "format": "prettier --write \"src/**/*.ts\"",
     "format:check": "prettier --check \"src/**/*.ts\"",
     "test": "vitest run",
+    "audit:lockfile": "node --experimental-strip-types scripts/check-lockfile-release-age.mts",
     "test:watch": "vitest",
     "changeset": "changeset",
     "release": "pnpm build && changeset publish"

package/tokenline.sh

@@ -56,6 +56,19 @@ epoch_from_iso() {
   fi
 }
 
+epoch_from_reset() {
+  # rate_limits.*.resets_at -> epoch seconds; empty on failure.
+  # Claude Code sends Unix epoch seconds, which `date` cannot parse as a date
+  # (GNU needs '@<n>', BSD needs '-r <n>'), so an integer is taken as-is and
+  # never reaches `date`. Anything else is tried as ISO-8601.
+  local value="$1"
+  if [[ "$value" =~ ^[0-9]+$ ]]; then
+    printf '%s' "$value"
+  elif [ -n "$value" ]; then
+    epoch_from_iso "$value"
+  fi
+}
+
 file_mtime() {
   local f="$1"
   if [ "$_stat_gnu" -eq 1 ]; then
@@ -95,7 +108,8 @@ parse_and_prepare_paths() {
     (.context_window.current_usage.input_tokens // 0),
     (.context_window.current_usage.output_tokens // 0),
     (.context_window.current_usage.cache_creation_input_tokens // 0),
-    (.context_window.current_usage.cache_read_input_tokens // 0)' 2>/dev/null)
+    (.context_window.current_usage.cache_read_input_tokens // 0),
+    (.model.id // "")' 2>/dev/null)
 
   # Malformed or empty stdin: jq emits nothing, so the array is empty. Degrade to
   # a silent no-op render rather than leaking parse errors or rendering garbage —
@@ -108,13 +122,14 @@ parse_and_prepare_paths() {
   transcript_path="${_f[3]}"
   session_id="${_f[4]}"
   rl_5h_pct="${_f[5]}"
-  rl_5h_reset=$(epoch_from_iso "${_f[6]}")
+  rl_5h_reset=$(epoch_from_reset "${_f[6]}")
   rl_7d_pct="${_f[7]}"
-  rl_7d_reset=$(epoch_from_iso "${_f[8]}")
+  rl_7d_reset=$(epoch_from_reset "${_f[8]}")
   cur_input="${_f[9]}"
   cur_output="${_f[10]}"
   cur_cwrite="${_f[11]}"
   cur_cread="${_f[12]}"
+  model_id="${_f[13]}"
 
   # Computed: total input-only tokens used in the current context window
   tokens_used=$((cur_input + cur_cwrite + cur_cread))
@@ -141,6 +156,15 @@ parse_and_prepare_paths() {
     is_gemini=true
   fi
 
+  # Claude Fable 5.1 and Mythos 5.1 bill cache hits at 0.025x base input instead of
+  # the standard 0.1x. Match the API id (claude-fable-5-1[1m]) or the display name
+  # (Fable 5.1) so either field alone is enough.
+  discounted_cache_read=false
+  local discounted_re='(fable|mythos)[- ]5[-.]1([^0-9]|$)'
+  if [[ "${model_id,,}" =~ $discounted_re ]] || [[ "${model,,}" =~ $discounted_re ]]; then
+    discounted_cache_read=true
+  fi
+
   # Get the current epoch timestamp once to be reused across all calculations
   now=$(date +%s)
 }
@@ -402,6 +426,7 @@ compute_turn_breakdown() {
       output_mult="4"
     else
       read_mult="0.1"
+      [ "$discounted_cache_read" = true ] && read_mult="0.025"
       write_mult="1.25"
       [ "${ttl_label:-5m}" = "1h" ] && write_mult="2"
       input_mult="1"