mirror of
https://github.com/tiennm99/ccs.git
synced 2026-07-16 20:17:45 +00:00
feat(profiles): add local runtime readiness checks
This commit is contained in:
@@ -64,6 +64,12 @@ export {
|
||||
// OpenRouter catalog and picker
|
||||
export { isOpenRouterUrl, fetchOpenRouterModels, type OpenRouterModel } from './openrouter-catalog';
|
||||
export { pickOpenRouterModel, type OpenRouterSelection } from './openrouter-picker';
|
||||
export {
|
||||
getLocalRuntimeReadiness,
|
||||
type LocalRuntimeId,
|
||||
type LocalRuntimeReadiness,
|
||||
type LocalRuntimeStatus,
|
||||
} from './local-runtime-readiness';
|
||||
|
||||
// Provider presets for CLI
|
||||
export {
|
||||
|
||||
@@ -0,0 +1,121 @@
|
||||
export type LocalRuntimeId = 'ollama' | 'llamacpp';
|
||||
export type LocalRuntimeStatus = 'ready' | 'missing-model' | 'offline';
|
||||
|
||||
export interface LocalRuntimeReadiness {
|
||||
id: LocalRuntimeId;
|
||||
name: string;
|
||||
endpoint: string;
|
||||
status: LocalRuntimeStatus;
|
||||
commandHint: string;
|
||||
recommendedModel: string | null;
|
||||
recommendedModelInstalled: boolean;
|
||||
detectedModelCount: number;
|
||||
}
|
||||
|
||||
interface LocalRuntimeDefinition {
|
||||
id: LocalRuntimeId;
|
||||
name: string;
|
||||
endpoint: string;
|
||||
modelsUrl: string;
|
||||
commandHint: string;
|
||||
recommendedModel: string | null;
|
||||
parseModelIds: (payload: unknown) => string[];
|
||||
}
|
||||
|
||||
const LOCAL_RUNTIME_DEFINITIONS: LocalRuntimeDefinition[] = [
|
||||
{
|
||||
id: 'ollama',
|
||||
name: 'Ollama',
|
||||
endpoint: 'http://127.0.0.1:11434',
|
||||
modelsUrl: 'http://127.0.0.1:11434/api/tags',
|
||||
commandHint: 'ollama serve',
|
||||
recommendedModel: 'gemma4:e4b',
|
||||
parseModelIds: (payload) => {
|
||||
const models = (payload as { models?: Array<{ name?: string }> })?.models;
|
||||
return Array.isArray(models)
|
||||
? models
|
||||
.map((model) => (typeof model?.name === 'string' ? model.name.trim() : ''))
|
||||
.filter(Boolean)
|
||||
: [];
|
||||
},
|
||||
},
|
||||
{
|
||||
id: 'llamacpp',
|
||||
name: 'llama.cpp',
|
||||
endpoint: 'http://127.0.0.1:8080',
|
||||
modelsUrl: 'http://127.0.0.1:8080/v1/models',
|
||||
commandHint: './server --host 0.0.0.0 --port 8080 -m model.gguf',
|
||||
recommendedModel: null,
|
||||
parseModelIds: (payload) => {
|
||||
const models = (payload as { data?: Array<{ id?: string }> })?.data;
|
||||
return Array.isArray(models)
|
||||
? models
|
||||
.map((model) => (typeof model?.id === 'string' ? model.id.trim() : ''))
|
||||
.filter(Boolean)
|
||||
: [];
|
||||
},
|
||||
},
|
||||
];
|
||||
|
||||
async function fetchModelIds(definition: LocalRuntimeDefinition): Promise<string[]> {
|
||||
const response = await fetch(definition.modelsUrl, {
|
||||
signal: AbortSignal.timeout(1500),
|
||||
});
|
||||
|
||||
if (!response.ok) {
|
||||
throw new Error(`${definition.id} readiness probe failed (${response.status})`);
|
||||
}
|
||||
|
||||
return definition.parseModelIds(await response.json());
|
||||
}
|
||||
|
||||
function toReadiness(
|
||||
definition: LocalRuntimeDefinition,
|
||||
modelIds: string[]
|
||||
): LocalRuntimeReadiness {
|
||||
const recommendedModelInstalled = definition.recommendedModel
|
||||
? modelIds.some((modelId) => modelId === definition.recommendedModel)
|
||||
: modelIds.length > 0;
|
||||
|
||||
const status: LocalRuntimeStatus =
|
||||
modelIds.length === 0 || !recommendedModelInstalled ? 'missing-model' : 'ready';
|
||||
|
||||
return {
|
||||
id: definition.id,
|
||||
name: definition.name,
|
||||
endpoint: definition.endpoint,
|
||||
status,
|
||||
commandHint:
|
||||
status === 'missing-model' && definition.recommendedModel
|
||||
? `ollama pull ${definition.recommendedModel}`
|
||||
: definition.commandHint,
|
||||
recommendedModel: definition.recommendedModel,
|
||||
recommendedModelInstalled,
|
||||
detectedModelCount: modelIds.length,
|
||||
};
|
||||
}
|
||||
|
||||
function toOffline(definition: LocalRuntimeDefinition): LocalRuntimeReadiness {
|
||||
return {
|
||||
id: definition.id,
|
||||
name: definition.name,
|
||||
endpoint: definition.endpoint,
|
||||
status: 'offline',
|
||||
commandHint: definition.commandHint,
|
||||
recommendedModel: definition.recommendedModel,
|
||||
recommendedModelInstalled: false,
|
||||
detectedModelCount: 0,
|
||||
};
|
||||
}
|
||||
|
||||
export async function getLocalRuntimeReadiness(): Promise<LocalRuntimeReadiness[]> {
|
||||
return Promise.all(
|
||||
LOCAL_RUNTIME_DEFINITIONS.map(async (definition) => {
|
||||
try {
|
||||
return toReadiness(definition, await fetchModelIds(definition));
|
||||
} catch {
|
||||
return toOffline(definition);
|
||||
}
|
||||
})
|
||||
);
|
||||
}
|
||||
@@ -10,6 +10,7 @@ import { isReservedName, RESERVED_PROFILE_NAMES } from '../../config/reserved-na
|
||||
import {
|
||||
createApiProfile,
|
||||
createCliproxyBridgeProfile,
|
||||
getLocalRuntimeReadiness,
|
||||
removeApiProfile,
|
||||
updateApiProfileTarget,
|
||||
discoverApiProfileOrphans,
|
||||
@@ -86,6 +87,14 @@ router.get('/cliproxy-bridge/providers', (_req: Request, res: Response): void =>
|
||||
}
|
||||
});
|
||||
|
||||
router.get('/local-runtime-readiness', async (_req: Request, res: Response): Promise<void> => {
|
||||
try {
|
||||
res.json({ runtimes: await getLocalRuntimeReadiness() });
|
||||
} catch (error) {
|
||||
res.status(500).json({ error: (error as Error).message });
|
||||
}
|
||||
});
|
||||
|
||||
router.post('/cliproxy-bridge', (req: Request, res: Response): void => {
|
||||
const shape = validatePayloadShape(req.body, ['provider', 'name', 'target']);
|
||||
if (!shape.ok) {
|
||||
|
||||
@@ -0,0 +1,141 @@
|
||||
import { afterAll, afterEach, beforeAll, beforeEach, describe, expect, it } from 'bun:test';
|
||||
import express from 'express';
|
||||
import type { Server } from 'http';
|
||||
import profileRoutes from '../../../src/web-server/routes/profile-routes';
|
||||
|
||||
describe('profile-routes local runtime readiness', () => {
|
||||
let server: Server;
|
||||
let baseUrl = '';
|
||||
const originalFetch = globalThis.fetch;
|
||||
|
||||
beforeAll(async () => {
|
||||
const app = express();
|
||||
app.use(express.json());
|
||||
app.use('/api/profiles', profileRoutes);
|
||||
|
||||
await new Promise<void>((resolve, reject) => {
|
||||
server = app.listen(0, '127.0.0.1');
|
||||
const onError = (error: Error) => reject(error);
|
||||
server.once('error', onError);
|
||||
server.once('listening', () => {
|
||||
server.off('error', onError);
|
||||
resolve();
|
||||
});
|
||||
});
|
||||
|
||||
const address = server.address();
|
||||
if (!address || typeof address === 'string') {
|
||||
throw new Error('Unable to resolve test server port');
|
||||
}
|
||||
|
||||
baseUrl = `http://127.0.0.1:${address.port}`;
|
||||
});
|
||||
|
||||
afterAll(async () => {
|
||||
await new Promise<void>((resolve) => server.close(() => resolve()));
|
||||
});
|
||||
|
||||
beforeEach(() => {
|
||||
globalThis.fetch = (async (input: RequestInfo | URL) => {
|
||||
const url = String(input);
|
||||
|
||||
if (url.startsWith(baseUrl)) {
|
||||
return originalFetch(input);
|
||||
}
|
||||
|
||||
if (url.includes('11434/api/tags')) {
|
||||
return new Response(
|
||||
JSON.stringify({
|
||||
models: [{ name: 'gemma4:e4b' }, { name: 'qwen3-coder:latest' }],
|
||||
}),
|
||||
{
|
||||
status: 200,
|
||||
headers: { 'Content-Type': 'application/json' },
|
||||
}
|
||||
);
|
||||
}
|
||||
|
||||
if (url.includes('8080/v1/models')) {
|
||||
return new Response(
|
||||
JSON.stringify({
|
||||
data: [{ id: 'Qwen3-Coder-30B-A3B-Instruct-Q4_K_M.gguf' }],
|
||||
}),
|
||||
{
|
||||
status: 200,
|
||||
headers: { 'Content-Type': 'application/json' },
|
||||
}
|
||||
);
|
||||
}
|
||||
|
||||
throw new Error(`Unexpected URL: ${url}`);
|
||||
}) as typeof fetch;
|
||||
});
|
||||
|
||||
afterEach(() => {
|
||||
globalThis.fetch = originalFetch;
|
||||
});
|
||||
|
||||
it('reports local runtimes as ready when their endpoints respond with models', async () => {
|
||||
const response = await fetch(`${baseUrl}/api/profiles/local-runtime-readiness`);
|
||||
|
||||
expect(response.status).toBe(200);
|
||||
const body = (await response.json()) as {
|
||||
runtimes: Array<{
|
||||
id: string;
|
||||
status: string;
|
||||
recommendedModelInstalled: boolean;
|
||||
}>;
|
||||
};
|
||||
|
||||
expect(body.runtimes).toHaveLength(2);
|
||||
expect(body.runtimes).toContainEqual(
|
||||
expect.objectContaining({
|
||||
id: 'ollama',
|
||||
status: 'ready',
|
||||
recommendedModelInstalled: true,
|
||||
})
|
||||
);
|
||||
expect(body.runtimes).toContainEqual(
|
||||
expect.objectContaining({
|
||||
id: 'llamacpp',
|
||||
status: 'ready',
|
||||
})
|
||||
);
|
||||
});
|
||||
|
||||
it('reports setup guidance when local endpoints are unavailable', async () => {
|
||||
globalThis.fetch = (async (input: RequestInfo | URL) => {
|
||||
const url = String(input);
|
||||
if (url.startsWith(baseUrl)) {
|
||||
return originalFetch(input);
|
||||
}
|
||||
throw new Error('connect ECONNREFUSED');
|
||||
}) as typeof fetch;
|
||||
|
||||
const response = await fetch(`${baseUrl}/api/profiles/local-runtime-readiness`);
|
||||
|
||||
expect(response.status).toBe(200);
|
||||
const body = (await response.json()) as {
|
||||
runtimes: Array<{
|
||||
id: string;
|
||||
status: string;
|
||||
commandHint: string;
|
||||
}>;
|
||||
};
|
||||
|
||||
expect(body.runtimes).toContainEqual(
|
||||
expect.objectContaining({
|
||||
id: 'ollama',
|
||||
status: 'offline',
|
||||
commandHint: 'ollama serve',
|
||||
})
|
||||
);
|
||||
expect(body.runtimes).toContainEqual(
|
||||
expect.objectContaining({
|
||||
id: 'llamacpp',
|
||||
status: 'offline',
|
||||
commandHint: './server --host 0.0.0.0 --port 8080 -m model.gguf',
|
||||
})
|
||||
);
|
||||
});
|
||||
});
|
||||
@@ -0,0 +1,11 @@
|
||||
import { useQuery } from '@tanstack/react-query';
|
||||
import { api } from '@/lib/api-client';
|
||||
|
||||
export function useLocalRuntimeReadiness() {
|
||||
return useQuery({
|
||||
queryKey: ['profiles', 'local-runtime-readiness'],
|
||||
queryFn: () => api.profiles.getLocalRuntimeReadiness(),
|
||||
staleTime: 15_000,
|
||||
refetchOnWindowFocus: false,
|
||||
});
|
||||
}
|
||||
@@ -239,6 +239,23 @@ export interface Profile {
|
||||
cliproxyBridge?: CliproxyBridgeMetadata | null;
|
||||
}
|
||||
|
||||
export type LocalRuntimeStatus = 'ready' | 'missing-model' | 'offline';
|
||||
|
||||
export interface LocalRuntimeReadiness {
|
||||
id: 'ollama' | 'llamacpp';
|
||||
name: string;
|
||||
endpoint: string;
|
||||
status: LocalRuntimeStatus;
|
||||
commandHint: string;
|
||||
recommendedModel: string | null;
|
||||
recommendedModelInstalled: boolean;
|
||||
detectedModelCount: number;
|
||||
}
|
||||
|
||||
export interface LocalRuntimeReadinessResponse {
|
||||
runtimes: LocalRuntimeReadiness[];
|
||||
}
|
||||
|
||||
export interface CreateProfile {
|
||||
name: string;
|
||||
baseUrl: string;
|
||||
@@ -1024,6 +1041,8 @@ export interface CliproxyRestartResult {
|
||||
export const api = {
|
||||
profiles: {
|
||||
list: () => request<{ profiles: Profile[] }>('/profiles'),
|
||||
getLocalRuntimeReadiness: () =>
|
||||
request<LocalRuntimeReadinessResponse>('/profiles/local-runtime-readiness'),
|
||||
create: (data: CreateProfile) =>
|
||||
request('/profiles', {
|
||||
method: 'POST',
|
||||
|
||||
Reference in New Issue
Block a user