token-monitor

trentuna/token-monitor

Fork 0

Commit graph

Author	SHA1	Message	Date
Vigilio Desto	c7e6438398	Fix xai probe: double /v1 URL bug, use /v1/models instead of chat completion Two bugs caused all xai providers to show 'error' in the monitor: 1. Double /v1 in URL: models.json baseUrl is https://api.x.ai/v1 (OpenAI- compatible convention), and the probe was appending /v1/chat/completions, producing https://api.x.ai/v1/v1/chat/completions → HTTP 4xx. Fix: strip trailing /vN from baseUrl before constructing the probe URL. 2. Wrong model: probe used grok-3-mini, which requires specific x.ai console permissions not granted to our keys. Keys have access to grok-4-1-fast-reasoning only. Fix: use GET /v1/models instead — lightweight, no model guessing, returns 200 (valid key) or 401 (invalid). Includes available models in result for visibility. 158/158 tests pass (unit tests for parseXaiHeaders unchanged).	2026-04-05 06:31:29 +00:00
Hannibal Smith	1b4e299461	build: add gemini and xai provider modules Expands token-monitor with two new provider types: - providers/gemini.js — Google Gemini API (body-based quota, no headers) - Probes generateContent endpoint (1 token), falls back gemini-2.0-flash → gemini-2.5-flash - Parses QuotaFailure violations + RetryInfo from 429 JSON body - Returns: status, quota_violations[], retry_delay_seconds, severity - providers/xai.js — x.ai/Grok (OpenAI-compatible header schema) - Reads x-ratelimit-{limit,remaining}-{requests,tokens} headers - Handles: no_key, ok, rate_limited, invalid_key states - Warning threshold: < 10% remaining on requests or tokens Both providers handle missing API keys gracefully (status: no_key). Classification via providers/index.js using baseUrl patterns. 140/140 tests passing. Closes recon findings from trentuna/a-team#91.	2026-04-04 17:52:37 +00:00

Author

SHA1

Message

Date

Vigilio Desto

c7e6438398

Fix xai probe: double /v1 URL bug, use /v1/models instead of chat completion

Two bugs caused all xai providers to show 'error' in the monitor:

1. Double /v1 in URL: models.json baseUrl is https://api.x.ai/v1 (OpenAI-
   compatible convention), and the probe was appending /v1/chat/completions,
   producing https://api.x.ai/v1/v1/chat/completions → HTTP 4xx.
   Fix: strip trailing /vN from baseUrl before constructing the probe URL.

2. Wrong model: probe used grok-3-mini, which requires specific x.ai
   console permissions not granted to our keys. Keys have access to
   grok-4-1-fast-reasoning only.
   Fix: use GET /v1/models instead — lightweight, no model guessing,
   returns 200 (valid key) or 401 (invalid). Includes available models
   in result for visibility.

158/158 tests pass (unit tests for parseXaiHeaders unchanged).

2026-04-05 06:31:29 +00:00

Hannibal Smith

1b4e299461

build: add gemini and xai provider modules

Expands token-monitor with two new provider types:

- providers/gemini.js — Google Gemini API (body-based quota, no headers)
  - Probes generateContent endpoint (1 token), falls back gemini-2.0-flash → gemini-2.5-flash
  - Parses QuotaFailure violations + RetryInfo from 429 JSON body
  - Returns: status, quota_violations[], retry_delay_seconds, severity

- providers/xai.js — x.ai/Grok (OpenAI-compatible header schema)
  - Reads x-ratelimit-{limit,remaining}-{requests,tokens} headers
  - Handles: no_key, ok, rate_limited, invalid_key states
  - Warning threshold: < 10% remaining on requests or tokens

Both providers handle missing API keys gracefully (status: no_key).
Classification via providers/index.js using baseUrl patterns.
140/140 tests passing.

Closes recon findings from trentuna/a-team#91.

2026-04-04 17:52:37 +00:00

2 commits