add adk and vertex ai integrations by johnymontana · Pull Request #35 · neo4j-labs/agent-memory

johnymontana · 2026-01-27T22:58:11Z

This PR implements integrations with Google Cloud's AI ecosystem for neo4j-agent-memory. It adds native support for Google's Agent Development Kit (ADK), Vertex AI embeddings, and exposes memory capabilities via MCP server for Cloud API Registry integration.

Features

1. Vertex AI Embeddings (VTX-001)

New VertexAIEmbedder class supporting:
- text-embedding-004 (768 dimensions, recommended)
- textembedding-gecko@003 (768 dimensions)
- textembedding-gecko-multilingual@001 (768 dimensions)
Batch embedding with 250 item limit (Vertex AI constraint)
Multiple authentication methods (ADC, service account, explicit credentials)
Task type support for retrieval and semantic similarity

2. Google ADK Integration (ADK-001)

Neo4jMemoryService implementing ADK's MemoryService interface
Session memory persistence with entity extraction
Semantic memory search across all memory types
Type conversion utilities between neo4j-agent-memory and ADK types

3. MCP Server (MCP-001)

Neo4jMemoryMCPServer with 5 tools for Cloud API Registry:
- memory_search - Hybrid vector + graph search
- memory_store - Store messages, facts, preferences
- entity_lookup - Get entity with graph neighbors
- conversation_history - Retrieve session messages
- graph_query - Execute read-only Cypher queries
Support for stdio and SSE transports
CLI command: neo4j-memory mcp serve

4. Cloud Run Template (RUN-001)

Production-ready Dockerfile
Cloud Build configuration
Terraform templates for infrastructure

5. Documentation & Diagrams

Comprehensive AsciiDoc documentation
5 Excalidraw architecture diagrams
Updated README with Google Cloud section

vercel · 2026-01-27T22:58:14Z

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
agent-memory	Ready	Preview, Comment	Feb 11, 2026 6:04am

Copilot

Pull request overview

This PR implements comprehensive Google Cloud integration for neo4j-agent-memory, adding native support for Vertex AI embeddings, Google's Agent Development Kit (ADK), and a Model Context Protocol (MCP) server for Cloud API Registry integration.

Changes:

Vertex AI Embeddings provider supporting text-embedding-004 and gecko models with 768-dimensional embeddings
Neo4jMemoryService implementing ADK's MemoryService interface for session persistence and semantic search
MCP server with 5 tools (memory_search, memory_store, entity_lookup, conversation_history, graph_query) supporting stdio and SSE transports
Cloud Run deployment templates with Dockerfile, Cloud Build configuration, and infrastructure setup
Comprehensive documentation and examples demonstrating all Google Cloud features

Reviewed changes

Copilot reviewed 41 out of 42 changed files in this pull request and generated 20 comments.

Show a summary per file

File	Description
`src/neo4j_agent_memory/embeddings/vertex_ai.py`	Vertex AI embedder implementation with batch processing
`src/neo4j_agent_memory/integrations/google_adk/`	ADK MemoryService and type conversions
`src/neo4j_agent_memory/mcp/`	MCP server, handlers, and tool definitions
`tests/unit/`	Comprehensive unit tests for all new components
`tests/integration/`	Integration tests with real Neo4j and Vertex AI
`examples/google_cloud_integration/`	Full-featured examples and demos
`deploy/cloudrun/`	Production deployment templates
`docs/`	Updated documentation with Google Cloud integration guide

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

src/neo4j_agent_memory/mcp/handlers.py

examples/google_cloud_integration/vertex_ai_embeddings.py

examples/google_cloud_integration/full_pipeline.py

deploy/cloudrun/Dockerfile

Copilot · 2026-01-27T23:52:48Z

src/neo4j_agent_memory/mcp/__init__.py

+"""
+
+__all__ = [
+    "Neo4jMemoryMCPServer",


The name 'Neo4jMemoryMCPServer' is exported by all but is not defined.

src/neo4j_agent_memory/mcp/__init__.py

Copilot · 2026-01-27T23:52:48Z

tests/unit/mcp/test_mcp_handlers.py

+"""Unit tests for MCP handlers."""
+
+import json
+from unittest.mock import AsyncMock, MagicMock, patch


Import of 'patch' is not used.

Copilot · 2026-01-27T23:52:48Z

tests/unit/mcp/test_mcp_tools.py

@@ -0,0 +1,164 @@
+"""Unit tests for MCP tool definitions."""
+
+import pytest


Import of 'pytest' is not used.

Copilot · 2026-01-27T23:52:49Z

tests/integration/test_vertex_ai_integration.py

+
+# Check if Vertex AI is available
+try:
+    import vertexai  # noqa: F401


Import of 'vertexai' is not used.

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

Copilot · 2026-02-05T14:34:48Z

@johnymontana I've opened a new pull request, #52, to work on those changes. Once the pull request is ready, I'll request review from you.

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

Copilot · 2026-02-05T14:35:34Z

@johnymontana I've opened a new pull request, #53, to work on those changes. Once the pull request is ready, I'll request review from you.

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

Copilot

Pull request overview

Copilot reviewed 98 out of 99 changed files in this pull request and generated 17 comments.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

src/neo4j_agent_memory/__init__.py

src/neo4j_agent_memory/embeddings/vertex_ai.py

tests/unit/embeddings/test_vertex_ai.py

examples/google-cloud-financial-advisor/frontend/src/components/Dashboard/CustomerDashboard.tsx

examples/google-cloud-financial-advisor/backend/src/api/routes/alerts.py

examples/google-cloud-financial-advisor/backend/src/api/routes/chat.py

examples/google-cloud-financial-advisor/backend/src/models/investigation.py

examples/google-cloud-financial-advisor/backend/src/api/routes/investigations.py

Copilot

Pull request overview

Copilot reviewed 117 out of 124 changed files in this pull request and generated 18 comments.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-02-09T04:09:59Z

src/neo4j_agent_memory/embeddings/vertex_ai.py

+        self._project_id = project_id
+        self._location = location
+        self._credentials = credentials
+        self._batch_size = min(batch_size, DEFAULT_BATCH_SIZE)


batch_size can be set to 0 or a negative number, which will make range(..., step=self._batch_size) raise at runtime. Consider validating batch_size >= 1 (raise a ValueError or clamp to 1) before using it.

Suggested change

self._batch_size = min(batch_size, DEFAULT_BATCH_SIZE)

self._batch_size = min(max(batch_size, 1), DEFAULT_BATCH_SIZE)

Copilot · 2026-02-09T04:10:00Z

src/neo4j_agent_memory/embeddings/vertex_ai.py

+            # Process in batches (Vertex AI limit is 250 per request)
+            for i in range(0, len(texts), self._batch_size):
+                batch = texts[i : i + self._batch_size]


batch_size can be set to 0 or a negative number, which will make range(..., step=self._batch_size) raise at runtime. Consider validating batch_size >= 1 (raise a ValueError or clamp to 1) before using it.

Suggested change

# Process in batches (Vertex AI limit is 250 per request)

for i in range(0, len(texts), self._batch_size):

batch = texts[i : i + self._batch_size]

# Determine and validate batch size

batch_size = self._batch_size

if batch_size is None or batch_size < 1:

raise EmbeddingError(f"Invalid batch_size {batch_size!r}; batch_size must be an integer >= 1.")

# Process in batches (Vertex AI limit is 250 per request)

for i in range(0, len(texts), batch_size):

batch = texts[i : i + batch_size]

Copilot · 2026-02-09T04:10:00Z

pyproject.toml

 anthropic = ["anthropic>=0.20.0"]
 sentence-transformers = ["sentence-transformers>=2.2.0"]
+vertex-ai = ["google-cloud-aiplatform>=1.38.0"]
+google = ["google-cloud-aiplatform>=1.38.0"]  # Alias for vertex-ai


The docs/examples install instructions use neo4j-agent-memory[google,mcp] to imply Vertex AI + ADK + MCP, but the google extra currently only installs google-cloud-aiplatform and does not include google-adk. Either (a) include google-adk (and any other required Google packages) in the google extra, or (b) update doc strings/examples to instruct installing [vertex-ai,google-adk,mcp] explicitly to avoid missing dependency surprises.

Suggested change

google = ["google-cloud-aiplatform>=1.38.0"] # Alias for vertex-ai

google = ["google-cloud-aiplatform>=1.38.0", "google-adk"] # Vertex AI + Google ADK

Copilot · 2026-02-09T04:10:00Z

pyproject.toml

+mcp = ["mcp>=1.0.0"]
+
+# Google ADK
+google-adk = ["google-adk>=0.1.0"]


The docs/examples install instructions use neo4j-agent-memory[google,mcp] to imply Vertex AI + ADK + MCP, but the google extra currently only installs google-cloud-aiplatform and does not include google-adk. Either (a) include google-adk (and any other required Google packages) in the google extra, or (b) update doc strings/examples to instruct installing [vertex-ai,google-adk,mcp] explicitly to avoid missing dependency surprises.

Copilot · 2026-02-09T04:10:00Z

docs/modules/ROOT/pages/how-to/integrations/index.adoc

+# Google Cloud (Vertex AI + ADK + MCP)
+pip install neo4j-agent-memory[google,mcp]


This install command currently doesn’t align with the package extras as defined in pyproject.toml (the google extra doesn’t include google-adk). Update the docs to match the actual extras, or adjust the extras so this command installs all components promised here.

Copilot · 2026-02-09T04:10:02Z

src/neo4j_agent_memory/mcp/tools.py

+    for tool in MEMORY_TOOLS:
+        if tool["name"] == name:
+            return tool


MEMORY_TOOLS.copy() is a shallow list copy, so callers can still mutate the inner dicts and accidentally affect global tool definitions; get_tool_by_name also returns the original dict. Consider returning deep copies (or immutable structures) to prevent surprising cross-call mutation.

Copilot · 2026-02-09T04:10:03Z

examples/google-cloud-financial-advisor/backend/src/api/routes/graph.py

+async def get_initialized_memory_service() -> FinancialMemoryService:
+    """Get the initialized memory service."""
+    service = get_memory_service()
+    if not service._initialized:
+        await service.initialize()
+    return service


This route relies on private attributes (service._initialized, client._driver, client._database). Since the library now exposes MemoryClient.graph for Neo4j access, prefer using that public API (e.g., memory_service.client.graph.execute_read(...)) and/or add a public readiness method on FinancialMemoryService to avoid tight coupling to internals that may change.

Copilot · 2026-02-09T04:10:03Z

examples/google-cloud-financial-advisor/backend/src/api/routes/graph.py

+        async with client._driver.session(database=client._database) as session:
+            result = await session.run(request.query, request.parameters)
+            records = await result.data()


This route relies on private attributes (service._initialized, client._driver, client._database). Since the library now exposes MemoryClient.graph for Neo4j access, prefer using that public API (e.g., memory_service.client.graph.execute_read(...)) and/or add a public readiness method on FinancialMemoryService to avoid tight coupling to internals that may change.

Copilot · 2026-02-09T04:10:03Z

examples/financial-services-advisor/data/load_sample_data.py

+    )
+
+    client = MemoryClient(settings)
+    await client.initialize()


MemoryClient usage elsewhere in this PR uses connect() / close() (and async with) rather than initialize(). If initialize() isn’t part of the public API, this script will fail at runtime. Consider switching to await client.connect() (or using async with MemoryClient(settings) as client:) to match the actual client lifecycle API.

Suggested change

await client.initialize()

await client.connect()

Copilot · 2026-02-09T04:10:03Z

examples/google-cloud-financial-advisor/docker-compose.yml

+      # Mount Google Cloud credentials for local development
+      - ${GOOGLE_APPLICATION_CREDENTIALS:-~/.config/gcloud/application_default_credentials.json}:/app/credentials.json:ro


Docker Compose does not reliably expand ~ in bind-mount paths, so the default value here can produce an invalid mount and break local startup. Prefer requiring an absolute path (or document/export it) and/or switch to ${HOME}/.config/... to make the default portable.

Suggested change

# Mount Google Cloud credentials for local development

- ${GOOGLE_APPLICATION_CREDENTIALS:-~/.config/gcloud/application_default_credentials.json}:/app/credentials.json:ro

# Mount Google Cloud credentials for local development.

# GOOGLE_APPLICATION_CREDENTIALS must be set to an absolute path on the host.

- ${GOOGLE_APPLICATION_CREDENTIALS:?GOOGLE_APPLICATION_CREDENTIALS must be set to an absolute path on the host}:/app/credentials.json:ro

Copilot

Pull request overview

Copilot reviewed 110 out of 130 changed files in this pull request and generated 11 comments.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-02-11T04:09:02Z

src/neo4j_agent_memory/embeddings/__init__.py

+from neo4j_agent_memory.embeddings.base import BaseEmbedder, Embedder
+from neo4j_agent_memory.embeddings.openai import OpenAIEmbedder


Importing OpenAIEmbedder unconditionally can break base installs where the OpenAI dependency is optional (ImportError when importing neo4j_agent_memory.embeddings). Consider making OpenAIEmbedder a lazy import via __getattr__ (like VertexAIEmbedder), or ensure OpenAI is a required dependency of the base package.

Copilot · 2026-02-11T04:09:02Z

src/neo4j_agent_memory/mcp/tools.py

+        "description": (
+            "Search across all memory types using hybrid vector + graph search. "
+            "Finds relevant messages, entities, preferences, and reasoning traces."
+        ),


The description says the search is across all memory types and includes reasoning traces, but the default memory_types omits traces. Either update the description to match the default behavior, or include traces in the default list.

Copilot · 2026-02-11T04:09:02Z

src/neo4j_agent_memory/mcp/tools.py

+                "memory_types": {
+                    "type": "array",
+                    "items": {
+                        "type": "string",
+                        "enum": ["messages", "entities", "preferences", "traces"],
+                    },
+                    "description": "Types of memory to search (defaults to all)",
+                    "default": ["messages", "entities", "preferences"],
+                },


The description says the search is across all memory types and includes reasoning traces, but the default memory_types omits traces. Either update the description to match the default behavior, or include traces in the default list.

Copilot · 2026-02-11T04:09:02Z

src/neo4j_agent_memory/mcp/tools.py

+    Returns:
+        List of tool definitions in MCP format.
+    """
+    return MEMORY_TOOLS.copy()


MEMORY_TOOLS.copy() is only a shallow copy of the list; callers can still mutate the underlying tool dicts (and impact future requests). Return a deep copy (e.g., via copy.deepcopy) or construct new dicts to keep tool definitions immutable for callers.

Copilot · 2026-02-11T04:09:03Z

src/neo4j_agent_memory/embeddings/vertex_ai.py

+        self._batch_size = min(batch_size, DEFAULT_BATCH_SIZE)
+        self._task_type = task_type
+        self._embedding_model: TextEmbeddingModel | None = None
+        self._initialized = False


self._initialized is set but not used anywhere in this class. Either remove it to reduce state/complexity, or use it consistently (e.g., to short-circuit initialization checks or expose an initialized property).

Copilot · 2026-02-11T04:09:03Z

deploy/cloudrun/service.yaml

+              valueFrom:
+                secretKeyRef:
+                  key: latest
+                  name: neo4j-user


This Cloud Run manifest requires a neo4j-user Secret Manager secret, but the repo’s setup instructions/scripts shown in this PR primarily create neo4j-uri and neo4j-password. Either update the setup docs/scripts to also create neo4j-user, or change the manifest to use a plain env var (common default is neo4j).

Suggested change

valueFrom:

secretKeyRef:

key: latest

name: neo4j-user

value: "neo4j"

Copilot · 2026-02-11T04:09:03Z

examples/google_cloud_integration/README.md

+pip install neo4j-agent-memory[google,mcp]
+
+# Or for development
+cd neo4j-agent-memory
+pip install -e ".[google,mcp]"


The README advertises installing "all Google Cloud features" with [google,mcp], but ADK support is exposed as a separate extra (google-adk) in pyproject.toml. Consider updating these commands to include google-adk (e.g., [google,google-adk,mcp]) so the documented installation actually matches the described features.

Suggested change

pip install neo4j-agent-memory[google,mcp]

# Or for development

cd neo4j-agent-memory

pip install -e ".[google,mcp]"

pip install neo4j-agent-memory[google,google-adk,mcp]

# Or for development

cd neo4j-agent-memory

pip install -e ".[google,google-adk,mcp]"

Copilot · 2026-02-11T04:09:04Z

examples/google-cloud-financial-advisor/backend/src/api/routes/graph.py

+    query_upper = request.query.upper().strip()
+    forbidden_keywords = [
+        "CREATE",
+        "MERGE",
+        "DELETE",
+        "REMOVE",
+        "SET",
+        "DROP",
+        "DETACH",
+        "CALL",
+        "LOAD",
+    ]
+
+    for keyword in forbidden_keywords:
+        if keyword in query_upper:
+            raise HTTPException(
+                status_code=400,
+                detail=f"Write operations ({keyword}) are not allowed. Use read-only queries.",
+            )


The validation uses substring matching, which will incorrectly reject legitimate read-only queries containing keywords inside identifiers (e.g., n.created_at contains CREATE after .upper()). Use a token-based check or a regex with word boundaries (and ideally ignore string literals/comments) to avoid false positives.

Copilot · 2026-02-11T04:09:04Z

examples/google-cloud-financial-advisor/backend/src/api/routes/graph.py

+        # Execute query through the memory client
+        client = memory_service.client
+        async with client._driver.session(database=client._database) as session:
+            result = await session.run(request.query, request.parameters)
+            records = await result.data()


This route uses private MemoryClient internals (_driver, _database). Since this PR introduces MemoryClient.graph for custom queries, prefer using that public API (e.g., client.graph.execute_read(...)) to avoid tight coupling to internal attributes that can change.

Suggested change

# Execute query through the memory client

client = memory_service.client

async with client._driver.session(database=client._database) as session:

result = await session.run(request.query, request.parameters)

records = await result.data()

# Execute query through the memory client's public graph interface

client = memory_service.client

records = await client.graph.execute_read(

request.query,

request.parameters,

)

Copilot · 2026-02-11T04:09:04Z

examples/google-cloud-financial-advisor/backend/src/api/routes/traces.py

+    if not service._initialized:
+        await service.initialize()


Accessing service._initialized relies on a private attribute. Prefer a public readiness check (or always call await service.initialize() since it is already idempotent) to avoid depending on internal state.

Suggested change

if not service._initialized:

await service.initialize()

await service.initialize()

add adk and vertex ai integrations

da1b5b6

vercel bot deployed to Preview January 27, 2026 22:58 View deployment

fix imports

463d05c

vercel bot deployed to Preview January 27, 2026 23:07 View deployment

fix - broken xref

a712f3f

vercel bot deployed to Preview January 27, 2026 23:37 View deployment

johnymontana marked this pull request as ready for review January 27, 2026 23:39

johnymontana requested a review from Copilot January 27, 2026 23:41

Copilot started reviewing on behalf of johnymontana January 27, 2026 23:42 View session

Copilot AI reviewed Jan 27, 2026

View reviewed changes

merge main

a4b47e5

vercel bot deployed to Preview February 5, 2026 02:09 View deployment

add ADK multi-agent example

99bca14

vercel bot deployed to Preview February 5, 2026 14:29 View deployment

johnymontana requested a review from Copilot February 5, 2026 14:32

Copilot started reviewing on behalf of johnymontana February 5, 2026 14:33 View session

Update examples/google_cloud_integration/full_pipeline.py

82e2a87

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

vercel bot deployed to Preview February 5, 2026 14:34 View deployment

Update src/neo4j_agent_memory/mcp/__init__.py

05e3091

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

vercel bot deployed to Preview February 5, 2026 14:34 View deployment

Update deploy/cloudrun/Dockerfile

ff59aea

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

Copilot AI mentioned this pull request Feb 5, 2026

Fix test mocks to use created_at instead of timestamp for Message objects #52

Open

vercel bot deployed to Preview February 5, 2026 14:35 View deployment

Copilot AI mentioned this pull request Feb 5, 2026

Expose model and task_type as public properties in VertexAIEmbedder #53

Open

Update deploy/cloudrun/Dockerfile

d232934

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

vercel bot deployed to Preview February 5, 2026 14:36 View deployment

Copilot AI reviewed Feb 5, 2026

View reviewed changes

vercel bot deployed to Preview February 8, 2026 23:58 View deployment

johnymontana added 3 commits February 8, 2026 17:04

update sample prompts

81cac20

update readme

118a561

fix async await

65839dc

vercel bot deployed to Preview February 9, 2026 00:10 View deployment

fixes

6ccb6e0

vercel bot deployed to Preview February 9, 2026 00:20 View deployment

johnymontana added 2 commits February 8, 2026 17:26

fix sample data import

1abae4d

update docs

22f700a

vercel bot deployed to Preview February 9, 2026 00:55 View deployment

add neo4j service

99d1832

vercel bot deployed to Preview February 9, 2026 01:31 View deployment

johnymontana added 2 commits February 8, 2026 18:52

update docs

c30ea2a

update roadmap

ef504a6

vercel bot deployed to Preview February 9, 2026 01:59 View deployment

add missing data files

fb17834

vercel bot deployed to Preview February 9, 2026 04:05 View deployment

johnymontana requested a review from Copilot February 9, 2026 04:06

Copilot AI reviewed Feb 9, 2026

View reviewed changes

improve agent orchestration view and save reasoning traces

b7921c7

vercel bot deployed to Preview February 9, 2026 06:28 View deployment

update docs and tests

acd46a3

vercel bot deployed to Preview February 9, 2026 07:30 View deployment

fix linting

1e54e9b

vercel bot deployed to Preview February 11, 2026 04:04 View deployment

johnymontana requested a review from Copilot February 11, 2026 04:05

Copilot AI reviewed Feb 11, 2026

View reviewed changes

cleanup

473cfb5

vercel bot deployed to Preview February 11, 2026 06:04 View deployment

johnymontana merged commit 6847272 into main Feb 11, 2026
20 checks passed

		@@ -0,0 +1,164 @@
		"""Unit tests for MCP tool definitions."""

		import pytest

	self._batch_size = min(batch_size, DEFAULT_BATCH_SIZE)
	self._batch_size = min(max(batch_size, 1), DEFAULT_BATCH_SIZE)

-            # Process in batches (Vertex AI limit is 250 per request)
-            for i in range(0, len(texts), self._batch_size):
-                batch = texts[i : i + self._batch_size]
+            # Determine and validate batch size
+            batch_size = self._batch_size
+            if batch_size is None or batch_size < 1:
+                raise EmbeddingError(f"Invalid batch_size {batch_size!r}; batch_size must be an integer >= 1.")
+            # Process in batches (Vertex AI limit is 250 per request)
+            for i in range(0, len(texts), batch_size):
+                batch = texts[i : i + batch_size]

	google = ["google-cloud-aiplatform>=1.38.0"] # Alias for vertex-ai
	google = ["google-cloud-aiplatform>=1.38.0", "google-adk"] # Vertex AI + Google ADK

		# Google Cloud (Vertex AI + ADK + MCP)
		pip install neo4j-agent-memory[google,mcp]

		# Mount Google Cloud credentials for local development
		- ${GOOGLE_APPLICATION_CREDENTIALS:-~/.config/gcloud/application_default_credentials.json}:/app/credentials.json:ro

-      # Mount Google Cloud credentials for local development
-      - ${GOOGLE_APPLICATION_CREDENTIALS:-~/.config/gcloud/application_default_credentials.json}:/app/credentials.json:ro
+      # Mount Google Cloud credentials for local development.
+      # GOOGLE_APPLICATION_CREDENTIALS must be set to an absolute path on the host.
+      - ${GOOGLE_APPLICATION_CREDENTIALS:?GOOGLE_APPLICATION_CREDENTIALS must be set to an absolute path on the host}:/app/credentials.json:ro

		from neo4j_agent_memory.embeddings.base import BaseEmbedder, Embedder
		from neo4j_agent_memory.embeddings.openai import OpenAIEmbedder

	if not service._initialized:
	await service.initialize()
	await service.initialize()

Comments

Conversation

johnymontana commented Jan 27, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Features

Uh oh!

vercel bot commented Jan 27, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Copilot AI Jan 27, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Copilot AI Jan 27, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 27, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 27, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI commented Feb 5, 2026

Uh oh!

Copilot AI commented Feb 5, 2026

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Copilot AI Feb 9, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 9, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 9, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 9, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 9, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 9, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 9, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 9, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 9, 2026

Choose a reason for hiding this comment

johnymontana commented Jan 27, 2026 •

edited

Loading

vercel bot commented Jan 27, 2026 •

edited

Loading