Evenfire is an intelligence layer platform for running LLM agents that take real actions, on your infrastructure
TypeScript
import { describe, expect, it } from 'vitest'
import { readFileSync, readdirSync, statSync } from 'node:fs'
import { join } from 'node:path'
const SRC_ROOT = new URL('../src', import.meta.url).pathname
const FORBIDDEN = [
'CLERUM_CONTROL_API',
'controlApiUrl',
'controlApiServiceName',
'controlApiServiceToken',
'nginx-workflow-approval-gateway',
'workflow-approval-gateway',
'x-service-token',
'/api/v1/internal/notifications',
'/api/v1/internal/workflow-approval-mediums',
]
function sourceFiles(dir: string): string[] {
return readdirSync(dir).flatMap(name => {
const path = join(dir, name)
const stat = statSync(path)
if (stat.isDirectory()) return sourceFiles(path)
return /\.(ts|tsx)$/.test(name) ? [path] : []
})
}
describe('channel-reader control-api boundary', () => {
it('keeps workflow approval delivery and verification behind mcp-host runtime APIs', () => {
const haystack = sourceFiles(SRC_ROOT)
.map(path => readFileSync(path, 'utf8'))
.join('\n')
for (const needle of FORBIDDEN) {
expect(haystack, `forbidden direct control-api dependency: ${needle}`).not.toContain(needle)
}
})
})