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 } from 'node:fs'
function readBaseControlPlaneConfigmaps(): string {
return readFileSync(
new URL('../../deploy/base/control-plane/configmaps.yaml', import.meta.url),
'utf-8'
)
}
function yamlDocs(content: string): string[] {
return content
.split(/\n---\n/)
.map(d => d.trim())
.filter(Boolean)
}
function docContaining(docs: string[], substring: string): string {
const hit = docs.find(d => d.includes(substring))
if (!hit) throw new Error(`No YAML document containing: ${substring}`)
return hit
}
function locationBlock(config: string, marker: string): string {
const start = config.indexOf(marker)
if (start < 0) throw new Error(`No gateway location containing: ${marker}`)
const next = config.indexOf('\n location ', start + marker.length)
return next < 0 ? config.slice(start) : config.slice(start, next)
}
describe('workflow approval medium gateway routes', () => {
it('exposes Telegram provider-event confirmations only through the mcp-host JWT gateway route', () => {
const gatewayConf = docContaining(
yamlDocs(readBaseControlPlaneConfigmaps()),
'name: nginx-workflow-approval-gateway'
)
expect(gatewayConf).not.toContain(
'/api/v1/internal/workflow-approval-mediums/telegram/challenges/confirm-provider-event'
)
const route = locationBlock(
gatewayConf,
'location = /api/v1/workflow-approval-mediums/telegram/challenges/confirm-provider-event'
)
expect(route).toContain('limit_except POST')
expect(route).toContain('proxy_set_header Authorization $http_authorization;')
expect(route).not.toContain('proxy_set_header x-service-token $http_x_service_token;')
expect(gatewayConf).not.toContain('/api/v1/admin/workflow-approval-mediums')
})
})