@cendor/tokenguard 0.2.2 → 0.2.4

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (3) hide show
  1. package/NOTICE +7 -0
  2. package/README.md +3 -3
  3. package/package.json +9 -3
package/NOTICE ADDED
@@ -0,0 +1,7 @@
1
+ PowerAI Labs
2
+ Copyright 2026 Raghav Mishra (PowerAI Labs)
3
+
4
+ This product includes software developed by Raghav Mishra (PowerAI Labs).
5
+
6
+ Licensed under the Apache License, Version 2.0. See the LICENSE file for the
7
+ full terms.
package/README.md CHANGED
@@ -1,7 +1,8 @@
1
1
  # @cendor/tokenguard
2
2
 
3
- Pre-flight cost caps + free per-tag spend attribution for LLM calls. The TypeScript port of
4
- `cendor.tokenguard`.
3
+ [![npm version](https://img.shields.io/npm/v/@cendor/tokenguard.svg)](https://www.npmjs.com/package/@cendor/tokenguard) [![License: Apache 2.0](https://img.shields.io/badge/License-Apache_2.0-blue.svg)](https://opensource.org/licenses/Apache-2.0)
4
+
5
+ Spending limits for LLM calls: stop a call before it runs if it would blow the budget, and see cost broken down per feature. The TypeScript port of `cendor.tokenguard`.
5
6
 
6
7
  tokenguard **subscribes** to `@cendor/core`'s event bus and registers a pre-flight interceptor — it
7
8
  never patches a client itself (the locked architecture: one `instrument()`, many subscribers). Once a
@@ -75,4 +76,3 @@ async, single-threaded TS runtime:
75
76
  **projections** use exact token counts. The OS-thread thread-safety tests are dropped (Node is
76
77
  single-threaded); `QueueSink` is an async FIFO drain loop rather than a daemon thread, preserving
77
78
  the observable semantics (FIFO order, back-pressure, idempotent close, flush→close ordering).
78
- ```
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@cendor/tokenguard",
3
- "version": "0.2.2",
3
+ "version": "0.2.4",
4
4
  "description": "Spending limits for LLM calls: stop a call before it runs if it would blow the budget, and see cost broken down per feature.",
5
5
  "type": "module",
6
6
  "license": "Apache-2.0",
@@ -18,7 +18,9 @@
18
18
  },
19
19
  "files": [
20
20
  "dist",
21
- "README.md"
21
+ "README.md",
22
+ "LICENSE",
23
+ "NOTICE"
22
24
  ],
23
25
  "sideEffects": [
24
26
  "./dist/index.js"
@@ -27,7 +29,7 @@
27
29
  "node": ">=18"
28
30
  },
29
31
  "dependencies": {
30
- "@cendor/core": "^0.3.1"
32
+ "@cendor/core": "^0.3.3"
31
33
  },
32
34
  "optionalDependencies": {
33
35
  "better-sqlite3": "^12.11.1"
@@ -52,6 +54,10 @@
52
54
  "url": "git+https://github.com/cendorhq/cendor-libs-js.git",
53
55
  "directory": "packages/tokenguard"
54
56
  },
57
+ "homepage": "https://cendor.ai/docs/tokenguard",
58
+ "bugs": {
59
+ "url": "https://github.com/cendorhq/cendor-libs-js/issues"
60
+ },
55
61
  "publishConfig": {
56
62
  "access": "public"
57
63
  }