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.
This commit is contained in:
parent
988618e165
commit
1b4e299461
2 changed files with 270 additions and 0 deletions
139
providers/xai.js
Normal file
139
providers/xai.js
Normal file
|
|
@ -0,0 +1,139 @@
|
|||
/**
|
||||
* xai.js — x.ai (Grok) provider
|
||||
*
|
||||
* OpenAI-compatible API. Rate-limit headers on authenticated responses:
|
||||
* x-ratelimit-limit-requests integer
|
||||
* x-ratelimit-remaining-requests integer
|
||||
* x-ratelimit-reset-requests ISO8601 or relative string
|
||||
* x-ratelimit-limit-tokens integer
|
||||
* x-ratelimit-remaining-tokens integer
|
||||
* x-ratelimit-reset-tokens ISO8601 or relative string
|
||||
*
|
||||
* Probe: POST /v1/chat/completions
|
||||
* Authorization: Bearer <key>
|
||||
* model: 'grok-3-mini'
|
||||
* max_tokens: 1
|
||||
*
|
||||
* Provider type: 'xai-api'
|
||||
*/
|
||||
|
||||
/**
|
||||
* Compute severity for an x.ai result object.
|
||||
*
|
||||
* @param {Object} result — partially-built result (status + limit/remaining fields)
|
||||
* @returns {'critical'|'warning'|'ok'|'unknown'}
|
||||
*/
|
||||
function xaiSeverity(result) {
|
||||
if (result.status === 'rate_limited') return 'critical';
|
||||
if (result.status === 'ok') {
|
||||
const reqPct = (result.requests_limit > 0)
|
||||
? result.requests_remaining / result.requests_limit
|
||||
: null;
|
||||
const tokPct = (result.tokens_limit > 0)
|
||||
? result.tokens_remaining / result.tokens_limit
|
||||
: null;
|
||||
if ((reqPct !== null && reqPct < 0.1) || (tokPct !== null && tokPct < 0.1)) return 'warning';
|
||||
return 'ok';
|
||||
}
|
||||
return 'unknown';
|
||||
}
|
||||
|
||||
/**
|
||||
* Parse x.ai rate-limit headers from a response.
|
||||
* Synchronous — suitable for unit tests with mock headers.
|
||||
*
|
||||
* @param {Object|null} headers — fetch Response.headers (or compatible mock with .get(name))
|
||||
* @param {number} httpStatus — HTTP status code
|
||||
* @param {string|null} apiKey — API key (null/undefined → no_key result)
|
||||
* @returns {Object} normalized provider result
|
||||
*/
|
||||
export function parseXaiHeaders(headers, httpStatus, apiKey) {
|
||||
if (!apiKey) {
|
||||
return { type: 'xai-api', status: 'no_key', severity: 'unknown' };
|
||||
}
|
||||
|
||||
if (httpStatus === 401) {
|
||||
return { type: 'xai-api', status: 'invalid_key', severity: 'unknown' };
|
||||
}
|
||||
|
||||
if (httpStatus === 429) {
|
||||
const remaining = headers
|
||||
? parseInt(headers.get('x-ratelimit-remaining-requests'), 10)
|
||||
: NaN;
|
||||
return {
|
||||
type: 'xai-api',
|
||||
status: 'rate_limited',
|
||||
requests_remaining: isNaN(remaining) ? 0 : remaining,
|
||||
severity: 'critical',
|
||||
};
|
||||
}
|
||||
|
||||
if (httpStatus === 200) {
|
||||
const h = (name) => (headers ? headers.get(name) : null);
|
||||
|
||||
const reqLimit = parseInt(h('x-ratelimit-limit-requests'), 10);
|
||||
const reqRemaining = parseInt(h('x-ratelimit-remaining-requests'), 10);
|
||||
const tokLimit = parseInt(h('x-ratelimit-limit-tokens'), 10);
|
||||
const tokRemaining = parseInt(h('x-ratelimit-remaining-tokens'), 10);
|
||||
|
||||
const result = {
|
||||
type: 'xai-api',
|
||||
status: 'ok',
|
||||
requests_limit: isNaN(reqLimit) ? null : reqLimit,
|
||||
requests_remaining: isNaN(reqRemaining) ? null : reqRemaining,
|
||||
tokens_limit: isNaN(tokLimit) ? null : tokLimit,
|
||||
tokens_remaining: isNaN(tokRemaining) ? null : tokRemaining,
|
||||
};
|
||||
|
||||
result.severity = xaiSeverity(result);
|
||||
return result;
|
||||
}
|
||||
|
||||
// Other status codes (403, 5xx, etc.)
|
||||
return {
|
||||
type: 'xai-api',
|
||||
status: 'error',
|
||||
message: `HTTP ${httpStatus}`,
|
||||
severity: 'unknown',
|
||||
};
|
||||
}
|
||||
|
||||
/**
|
||||
* Probe an x.ai API endpoint.
|
||||
*
|
||||
* @param {string} providerName
|
||||
* @param {string} baseUrl — base URL (e.g. https://api.x.ai)
|
||||
* @param {string|null} apiKey — x.ai API key
|
||||
* @returns {Promise<Object>} normalized provider result
|
||||
*/
|
||||
export async function probeXaiProvider(providerName, baseUrl, apiKey) {
|
||||
if (!apiKey) {
|
||||
return { type: 'xai-api', status: 'no_key', severity: 'unknown' };
|
||||
}
|
||||
|
||||
const base = (baseUrl || 'https://api.x.ai').replace(/\/$/, '');
|
||||
|
||||
try {
|
||||
const response = await fetch(`${base}/v1/chat/completions`, {
|
||||
method: 'POST',
|
||||
headers: {
|
||||
'Authorization': `Bearer ${apiKey}`,
|
||||
'content-type': 'application/json',
|
||||
},
|
||||
body: JSON.stringify({
|
||||
model: 'grok-3-mini',
|
||||
max_tokens: 1,
|
||||
messages: [{ role: 'user', content: 'Hi' }],
|
||||
}),
|
||||
});
|
||||
|
||||
return parseXaiHeaders(response.headers, response.status, apiKey);
|
||||
} catch (err) {
|
||||
return {
|
||||
type: 'xai-api',
|
||||
status: 'error',
|
||||
message: err.message,
|
||||
severity: 'unknown',
|
||||
};
|
||||
}
|
||||
}
|
||||
Loading…
Add table
Add a link
Reference in a new issue