426e9eeabd
Benchmark Bridge Tests / benchmark (bunx @biomejs/biome check packages/lifeops-bench/src, benchmark-lint) (push) Waiting to run
Benchmark Bridge Tests / benchmark (bunx vitest run --config packages/lifeops-bench/vitest.config.ts --root packages/lifeops-bench --passWithNoTests, benchmark-tests) (push) Waiting to run
Build Agent Image / build-and-push (push) Waiting to run
Chat shell gestures / Chat shell gesture + parity e2e (push) Waiting to run
Cloud Gateway Discord / Test (push) Waiting to run
Cloud Gateway Webhook / Test (push) Waiting to run
Cloud Tests / lint-and-types (push) Waiting to run
Cloud Tests / unit-tests (push) Waiting to run
Cloud Tests / integration-tests (push) Waiting to run
Cloud Tests / e2e-tests (push) Blocked by required conditions
CodeQL Advanced / Analyze (javascript-typescript) (push) Waiting to run
Deploy Apps Worker (Product 2) / Determine environment (push) Waiting to run
Deploy Apps Worker (Product 2) / Deploy apps worker to apps-control host (${{ needs.determine-env.outputs.environment }}) (push) Blocked by required conditions
Deploy Eliza Provisioning Worker / Determine environment (push) Waiting to run
Deploy Eliza Provisioning Worker / Deploy worker to Hetzner host (${{ needs.determine-env.outputs.environment }} @ ${{ needs.determine-env.outputs.deployment_sha }}) (push) Blocked by required conditions
Dev Smoke / Classify changed paths (push) Waiting to run
Dev Smoke / bun run dev onboarding chat (push) Blocked by required conditions
Dev Smoke / Vite HMR dependency-level smoke (push) Blocked by required conditions
Electrobun Submodule Guard / electrobun gitlink is fetchable (push) Waiting to run
Publish @elizaos/example-code / check_npm (push) Waiting to run
Publish @elizaos/example-code / publish_npm (push) Blocked by required conditions
Publish @elizaos/plugin-elizacloud / verify_version (push) Waiting to run
Publish @elizaos/plugin-elizacloud / publish_npm (push) Blocked by required conditions
Sandbox Live Smoke / Sandbox live smoke (push) Waiting to run
Snap Build & Test / Build Snap (amd64) (push) Waiting to run
Snap Build & Test / Build Snap (arm64) (push) Waiting to run
supply-chain / sbom (push) Waiting to run
supply-chain / vulnerability-scan (push) Waiting to run
Build, Push & Deploy to Phala Cloud / build-and-push (push) Waiting to run
Test Packaging / Validate Packaging Configs (push) Waiting to run
Test Packaging / PyPI on Python ${{ matrix.python }} (push) Waiting to run
Test Packaging / Pack & Test JS Tarballs (push) Waiting to run
Test Packaging / elizaos CLI global-install smoke (node + bun) (push) Waiting to run
UI Fixture E2E / ui-fixture-e2e (push) Waiting to run
UI Fixture E2E / fixture-e2e (push) Waiting to run
UI Story Gate / story-gate (push) Waiting to run
vault-ci / test (macos-latest) (push) Waiting to run
vault-ci / test (ubuntu-latest) (push) Waiting to run
vault-ci / test (windows-latest) (push) Waiting to run
vault-ci / app-core wiring tests (push) Waiting to run
verify-patches / verify patches/CHECKSUMS.sha256 (push) Waiting to run
Voice Benchmark Smoke / voice-emotion fixture smoke (push) Waiting to run
Voice Benchmark Smoke / voiceagentbench fixture smoke (push) Waiting to run
Voice Benchmark Smoke / voicebench-quality unit smoke (push) Waiting to run
Voice Benchmark Smoke / voicebench TypeScript unit (no audio) (push) Waiting to run
Voice Benchmark Smoke / voice bench smoke summary (push) Blocked by required conditions
Windows CI / windows ([bun run --cwd packages/app-core test bun run --cwd packages/elizaos test bun run --cwd packages/cloud/shared test], app-and-cli) (push) Waiting to run
Windows CI / windows ([bun run --cwd packages/scenario-runner test bun run --cwd packages/vault test bun run --cwd packages/security test bun run --cwd plugins/plugin-coding-tools test], framework-packages) (push) Waiting to run
Windows CI / windows ([bun run --cwd plugins/plugin-elizacloud test bun run --cwd plugins/plugin-discord test bun run --cwd plugins/plugin-anthropic test bun run --cwd plugins/plugin-openai test bun run --cwd plugins/plugin-app-control test bun run --cwd plugins/pl… (push) Waiting to run
Windows CI / windows ([node packages/scripts/run-turbo.mjs run build --filter=@elizaos/core --filter=@elizaos/shared --filter=@elizaos/agent --concurrency=4 node packages/scripts/run-bash-linux-only.mjs scripts/verify-riscv64-buildpaths.sh node packages/scripts/run… (push) Waiting to run
Windows CI / windows ([node packages/scripts/run-turbo.mjs run typecheck --filter=@elizaos/core --filter=@elizaos/shared --filter=@elizaos/cloud-shared --concurrency=4 bun run --cwd packages/core test bun run --cwd packages/shared test], core-runtime, 75) (push) Waiting to run
Test Packaging / Build & Test PyPI Package (push) Waiting to run
Voice Workbench / headless workbench (mocked backends) (push) Has been cancelled
Voice Workbench / real acoustic lane (nightly, provisioned only) (push) Has been cancelled
ci / test (push) Has been cancelled
ci / lint-and-format (push) Has been cancelled
ci / build (push) Has been cancelled
ci / dev-startup (push) Has been cancelled
gitleaks / gitleaks (push) Has been cancelled
Markdown Links / Relative Markdown Links (push) Has been cancelled
Quality (Extended) / Homepage Build (PR smoke) (push) Has been cancelled
Quality (Extended) / Comment-only diff guard (push) Has been cancelled
Quality (Extended) / Format + Type Safety Ratchet (push) Has been cancelled
Quality (Extended) / Develop Gate (secret scan + UI determinism) (push) Has been cancelled
Quality (Extended) / Develop Gate (lint) (push) Has been cancelled
351 lines
9.7 KiB
TypeScript
351 lines
9.7 KiB
TypeScript
/**
|
|
* Live E2E tests for document integration.
|
|
*
|
|
* These tests use a real runtime, real embeddings, and a real LLM-backed chat
|
|
* route for retrieval.
|
|
*/
|
|
import path from "node:path";
|
|
import { createElizaPlugin } from "@elizaos/agent";
|
|
import { documentsPlugin } from "@elizaos/plugin-documents";
|
|
import { afterAll, beforeAll, expect, it } from "vitest";
|
|
import { describeIf } from "../../../packages/test/helpers/conditional-tests.ts";
|
|
import {
|
|
createConversation,
|
|
postConversationMessage,
|
|
req,
|
|
} from "../../../packages/test/helpers/http";
|
|
import {
|
|
isLiveTestEnabled,
|
|
selectLiveProvider,
|
|
} from "../../../packages/test/helpers/live-provider";
|
|
import { createRealTestRuntime } from "../../../packages/test/helpers/real-runtime.ts";
|
|
|
|
const envPath = path.resolve(
|
|
import.meta.dirname,
|
|
"..",
|
|
"..",
|
|
"..",
|
|
"..",
|
|
".env",
|
|
);
|
|
try {
|
|
const { config } = await import("dotenv");
|
|
config({ path: envPath });
|
|
} catch {
|
|
// dotenv may not be available.
|
|
}
|
|
|
|
const LIVE_PROVIDER =
|
|
selectLiveProvider("cerebras") ??
|
|
selectLiveProvider("openai") ??
|
|
selectLiveProvider();
|
|
const CAN_RUN = isLiveTestEnabled() && Boolean(LIVE_PROVIDER);
|
|
const DOCUMENT_CODEWORD = "VELVET-MOON-4821";
|
|
|
|
type StartedDocumentServer = {
|
|
close: () => Promise<void>;
|
|
port: number;
|
|
};
|
|
|
|
async function askForExactDocumentCodeword(
|
|
port: number,
|
|
conversationId: string,
|
|
): Promise<string> {
|
|
const prompts = [
|
|
"What is the exact deployment codeword from the uploaded document? Reply with only the exact codeword, including hyphens and digits. Do not execute any actions.",
|
|
"Search the uploaded document and copy the exact string from the sentence 'The deployment codeword is ...'. Reply with the codeword only. Do not update any profile or entity.",
|
|
"Return only the exact uppercase-and-hyphen codeword from documents. No actions. No explanation.",
|
|
];
|
|
|
|
let lastText = "";
|
|
for (const text of prompts) {
|
|
const { status, data } = await postConversationMessage(
|
|
port,
|
|
conversationId,
|
|
{ text },
|
|
undefined,
|
|
{ timeoutMs: 120_000 },
|
|
);
|
|
|
|
expect(status).toBe(200);
|
|
lastText = String(data.text ?? data.response ?? "");
|
|
expect(lastText.length).toBeGreaterThan(0);
|
|
if (lastText.includes(DOCUMENT_CODEWORD)) {
|
|
return lastText;
|
|
}
|
|
}
|
|
|
|
return lastText;
|
|
}
|
|
|
|
async function startDocumentServer(): Promise<StartedDocumentServer> {
|
|
const runtimeResult = await createRealTestRuntime({
|
|
withLLM: true,
|
|
preferredProvider: LIVE_PROVIDER?.name,
|
|
plugins: [createElizaPlugin({ agentId: "main" }), documentsPlugin],
|
|
});
|
|
const { startApiServer } = await import("@elizaos/agent");
|
|
const server = await startApiServer({
|
|
port: 0,
|
|
runtime: runtimeResult.runtime,
|
|
skipDeferredStartupWork: true,
|
|
});
|
|
await req(server.port, "POST", "/api/agent/start");
|
|
|
|
return {
|
|
port: server.port,
|
|
close: async () => {
|
|
await server.close();
|
|
await runtimeResult.cleanup();
|
|
},
|
|
};
|
|
}
|
|
|
|
describeIf(CAN_RUN)("Live: document management flow", () => {
|
|
let server: StartedDocumentServer | null = null;
|
|
let uploadedDocumentId: string | null = null;
|
|
|
|
beforeAll(async () => {
|
|
server = await startDocumentServer();
|
|
}, 120_000);
|
|
|
|
afterAll(async () => {
|
|
await server?.close();
|
|
});
|
|
|
|
it("step 1: gets document stats", async () => {
|
|
const { status, data } = await req(
|
|
server?.port ?? 0,
|
|
"GET",
|
|
"/api/documents/stats",
|
|
);
|
|
expect(status).toBe(200);
|
|
expect(typeof data.documentCount).toBe("number");
|
|
expect(typeof data.fragmentCount).toBe("number");
|
|
expect(typeof data.agentId).toBe("string");
|
|
});
|
|
|
|
it("step 2: uploads a text document", async () => {
|
|
const testContent = `
|
|
# Test Document
|
|
|
|
The deployment codeword is ${DOCUMENT_CODEWORD}.
|
|
|
|
This document verifies that document upload, semantic search, and chat
|
|
retrieval all use the real runtime path.
|
|
|
|
RAG retrieval should answer questions about this codeword from the document.
|
|
`.trim();
|
|
|
|
const { status, data } = await req(
|
|
server?.port ?? 0,
|
|
"POST",
|
|
"/api/documents",
|
|
{
|
|
content: testContent,
|
|
filename: "test-document-doc.md",
|
|
contentType: "text/markdown",
|
|
},
|
|
);
|
|
|
|
expect(status).toBe(200);
|
|
expect(data.ok).toBe(true);
|
|
expect(typeof data.documentId).toBe("string");
|
|
expect(typeof data.fragmentCount).toBe("number");
|
|
expect(data.fragmentCount).toBeGreaterThan(0);
|
|
uploadedDocumentId = data.documentId as string;
|
|
});
|
|
|
|
it("step 3: lists documents including the uploaded doc", async () => {
|
|
const { status, data } = await req(
|
|
server?.port ?? 0,
|
|
"GET",
|
|
"/api/documents",
|
|
);
|
|
expect(status).toBe(200);
|
|
expect(Array.isArray(data.documents)).toBe(true);
|
|
|
|
const docs = data.documents as Array<{ filename: string; id: string }>;
|
|
const uploadedDoc = docs.find((entry) => entry.id === uploadedDocumentId);
|
|
expect(uploadedDoc?.filename).toContain("test-document-doc");
|
|
});
|
|
|
|
it("step 4: gets document details", async () => {
|
|
const { status, data } = await req(
|
|
server?.port ?? 0,
|
|
"GET",
|
|
`/api/documents/${encodeURIComponent(uploadedDocumentId ?? "")}`,
|
|
);
|
|
|
|
expect(status).toBe(200);
|
|
expect(data.document).toBeDefined();
|
|
const doc = data.document as {
|
|
contentType: string;
|
|
filename: string;
|
|
id: string;
|
|
};
|
|
expect(doc.id).toBe(uploadedDocumentId);
|
|
expect(doc.filename).toContain("test-document-doc");
|
|
expect(doc.contentType).toBe("text/markdown");
|
|
});
|
|
|
|
it("step 5: gets document fragments", async () => {
|
|
const { status, data } = await req(
|
|
server?.port ?? 0,
|
|
"GET",
|
|
`/api/documents/${encodeURIComponent(uploadedDocumentId ?? "")}/fragments`,
|
|
);
|
|
|
|
expect(status).toBe(200);
|
|
expect(data.documentId).toBe(uploadedDocumentId);
|
|
expect(Array.isArray(data.fragments)).toBe(true);
|
|
|
|
const fragments = data.fragments as Array<{ text: string }>;
|
|
expect(fragments.length).toBeGreaterThan(0);
|
|
expect(
|
|
fragments.some((fragment) => fragment.text.includes(DOCUMENT_CODEWORD)),
|
|
).toBe(true);
|
|
});
|
|
|
|
it("step 6: searches documents with semantic matching", async () => {
|
|
const { status, data } = await req(
|
|
server?.port ?? 0,
|
|
"GET",
|
|
"/api/documents/search?q=deployment%20codeword&threshold=0.2&limit=5",
|
|
);
|
|
|
|
expect(status).toBe(200);
|
|
expect(Array.isArray(data.results)).toBe(true);
|
|
const results = data.results as Array<{
|
|
similarity: number;
|
|
text: string;
|
|
}>;
|
|
expect(results.length).toBeGreaterThan(0);
|
|
expect(
|
|
results.some((result) => result.text.includes(DOCUMENT_CODEWORD)),
|
|
).toBe(true);
|
|
for (let index = 1; index < results.length; index += 1) {
|
|
expect(results[index].similarity).toBeLessThanOrEqual(
|
|
results[index - 1].similarity,
|
|
);
|
|
}
|
|
});
|
|
|
|
it("step 7: chat retrieves the uploaded document through the real runtime", async () => {
|
|
const conversation = await createConversation(server?.port ?? 0, {
|
|
title: "Document retrieval",
|
|
});
|
|
const text = await askForExactDocumentCodeword(
|
|
server?.port ?? 0,
|
|
conversation.conversationId,
|
|
);
|
|
expect(text).toContain(DOCUMENT_CODEWORD);
|
|
}, 120_000);
|
|
|
|
it("step 8: deletes document and fragments", async () => {
|
|
const { status, data } = await req(
|
|
server?.port ?? 0,
|
|
"DELETE",
|
|
`/api/documents/${encodeURIComponent(uploadedDocumentId ?? "")}`,
|
|
);
|
|
|
|
expect(status).toBe(200);
|
|
expect(data.ok).toBe(true);
|
|
expect(typeof data.deletedFragments).toBe("number");
|
|
expect(data.deletedFragments).toBeGreaterThan(0);
|
|
|
|
const { data: listData } = await req(
|
|
server?.port ?? 0,
|
|
"GET",
|
|
"/api/documents",
|
|
);
|
|
const docs = listData.documents as Array<{ id: string }>;
|
|
expect(docs.some((doc) => doc.id === uploadedDocumentId)).toBe(false);
|
|
});
|
|
});
|
|
|
|
describeIf(CAN_RUN)("Live: URL import validation", () => {
|
|
let server: StartedDocumentServer | null = null;
|
|
|
|
beforeAll(async () => {
|
|
server = await startDocumentServer();
|
|
}, 120_000);
|
|
|
|
afterAll(async () => {
|
|
await server?.close();
|
|
});
|
|
|
|
it("validates URL format", async () => {
|
|
const { status, data } = await req(
|
|
server?.port ?? 0,
|
|
"POST",
|
|
"/api/documents/url",
|
|
{
|
|
url: "not-a-valid-url",
|
|
},
|
|
);
|
|
expect(status).toBe(400);
|
|
expect(data.error).toContain("Invalid URL");
|
|
});
|
|
|
|
it("handles missing URL", async () => {
|
|
const { status, data } = await req(
|
|
server?.port ?? 0,
|
|
"POST",
|
|
"/api/documents/url",
|
|
{},
|
|
);
|
|
expect(status).toBe(400);
|
|
expect(data.error).toContain("url is required");
|
|
});
|
|
});
|
|
|
|
describeIf(CAN_RUN)("Live: empty document store behavior", () => {
|
|
let server: StartedDocumentServer | null = null;
|
|
|
|
beforeAll(async () => {
|
|
server = await startDocumentServer();
|
|
}, 120_000);
|
|
|
|
afterAll(async () => {
|
|
await server?.close();
|
|
});
|
|
|
|
it("document stats work when empty", async () => {
|
|
const { data: listData } = await req(
|
|
server?.port ?? 0,
|
|
"GET",
|
|
"/api/documents",
|
|
);
|
|
const docs = listData.documents as Array<{ id: string }>;
|
|
for (const doc of docs) {
|
|
await req(
|
|
server?.port ?? 0,
|
|
"DELETE",
|
|
`/api/documents/${encodeURIComponent(doc.id)}`,
|
|
);
|
|
}
|
|
|
|
const { status, data } = await req(
|
|
server?.port ?? 0,
|
|
"GET",
|
|
"/api/documents/stats",
|
|
);
|
|
expect(status).toBe(200);
|
|
expect(data.documentCount).toBe(0);
|
|
expect(data.fragmentCount).toBe(0);
|
|
});
|
|
|
|
it("search returns an empty array when no documents exist", async () => {
|
|
const { status, data } = await req(
|
|
server?.port ?? 0,
|
|
"GET",
|
|
"/api/documents/search?q=test%20query&threshold=0.3",
|
|
);
|
|
|
|
expect(status).toBe(200);
|
|
expect(Array.isArray(data.results)).toBe(true);
|
|
expect(data.results).toHaveLength(0);
|
|
});
|
|
});
|