{agentDetails?.permission_type.replace(/\s*\([^)]*\)/g, '') || ''}
diff --git a/superagi/controllers/agent.py b/superagi/controllers/agent.py
index d7852dada..0d960f12c 100644
--- a/superagi/controllers/agent.py
+++ b/superagi/controllers/agent.py
@@ -30,6 +30,7 @@
import json
from superagi.models.toolkit import Toolkit
+from superagi.models.knowledges import Knowledges
from sqlalchemy import func
# from superagi.types.db import AgentOut, AgentIn
@@ -480,13 +481,16 @@ def get_agent_configuration(agent_id: int,
AgentExecution.agent_id == agent_id).scalar()
total_tokens = db.session.query(func.sum(AgentExecution.num_of_tokens)).filter(
AgentExecution.agent_id == agent_id).scalar()
+
# Construct the JSON response
response = {result.key: result.value for result in results}
+ knowledge = db.session.query(Knowledges).filter(Knowledges.id == response['knowledge']).first()
response = merge(response, {"name": agent.name, "description": agent.description,
# Query the AgentConfiguration table for the speci
"goal": eval(response["goal"]),
"instruction": eval(response.get("instruction", '[]')),
+ "knowledge_name": knowledge.name,
"calls": total_calls,
"tokens": total_tokens,
"constraints": eval(response.get("constraints")),
diff --git a/superagi/controllers/knowledges.py b/superagi/controllers/knowledges.py
index dbafbf6db..3afac761f 100644
--- a/superagi/controllers/knowledges.py
+++ b/superagi/controllers/knowledges.py
@@ -126,7 +126,7 @@ def install_selected_knowledge(knowledge_name: str, vector_db_index_id: int, org
file_chunks = S3Helper().get_json_file(selected_knowledge_config["file_path"])
vector = Vectordbs.get_vector_db_from_id(db.session, vector_db_index.vector_db_id)
db_creds = VectordbConfigs.get_vector_db_config_from_db_id(db.session, vector.id)
- upsert_data = VectorEmbeddingFactory.build_vector_storge(vector.db_type).get_vector_embeddings_from_chunks(file_chunks)
+ upsert_data = VectorEmbeddingFactory.build_vector_storage(vector.db_type, file_chunks).get_vector_embeddings_from_chunks()
try:
vector_db_storage = VectorFactory.build_vector_storage(vector.db_type, vector_db_index.name, **db_creds)
vector_db_storage.add_embeddings_to_vector_db(upsert_data)
diff --git a/superagi/models/agent.py b/superagi/models/agent.py
index 719410545..f979e2442 100644
--- a/superagi/models/agent.py
+++ b/superagi/models/agent.py
@@ -106,9 +106,9 @@ def eval_agent_config(cls, key, value):
"""
- if key in ["name", "description", "agent_type", "exit", "model", "permission_type", "LTM_DB", "resource_summary"]:
+ if key in ["name", "description", "agent_type", "exit", "model", "permission_type", "LTM_DB", "resource_summary", "knowledge"]:
return value
- elif key in ["project_id", "memory_window", "max_iterations", "iteration_interval", "knowledge"]:
+ elif key in ["project_id", "memory_window", "max_iterations", "iteration_interval"]:
return int(value)
elif key in ["goal", "constraints", "instruction", "is_deleted"]:
return eval(value)
diff --git a/superagi/vector_embeddings/vector_embedding_factory.py b/superagi/vector_embeddings/vector_embedding_factory.py
index 68beab835..827edb79c 100644
--- a/superagi/vector_embeddings/vector_embedding_factory.py
+++ b/superagi/vector_embeddings/vector_embedding_factory.py
@@ -9,7 +9,7 @@
class VectorEmbeddingFactory:
@classmethod
- def build_vector_storge(cls, vector_store: VectorStoreType, chunk_json: Optional[dict] = None):
+ def build_vector_storage(cls, vector_store: VectorStoreType, chunk_json: Optional[dict] = None):
"""
Get the vector embeddings from final chunks.
Args:
@@ -18,14 +18,14 @@ def build_vector_storge(cls, vector_store: VectorStoreType, chunk_json: Optional
The vector storage object
"""
final_chunks = []
+ uuid = []
+ embeds = []
+ metadata = []
vector_store = VectorStoreType.get_vector_store_type(vector_store)
if chunk_json is not None:
for key in chunk_json.keys():
final_chunks.append(chunk_json[key])
- uuid = []
- embeds = []
- metadata = []
for i in range(0, len(final_chunks)):
uuid.append(final_chunks[i]["id"])
embeds.append(final_chunks[i]["embeds"])
@@ -35,6 +35,7 @@ def build_vector_storge(cls, vector_store: VectorStoreType, chunk_json: Optional
'knowledge_name': final_chunks[i]['knowledge_name']
}
metadata.append(data)
+
if vector_store == VectorStoreType.PINECONE:
return Pinecone(uuid, embeds, metadata)
diff --git a/tests/unit_tests/vector_embeddings/test_vector_embedding_factory.py b/tests/unit_tests/vector_embeddings/test_vector_embedding_factory.py
index b381e208d..de447c9c6 100644
--- a/tests/unit_tests/vector_embeddings/test_vector_embedding_factory.py
+++ b/tests/unit_tests/vector_embeddings/test_vector_embedding_factory.py
@@ -12,13 +12,13 @@ def test_build_vector_storge(self, mock_qdrant, mock_pinecone):
"2": {"id": 2, "embeds": [4,5,6], "text": "test2", "chunk": "chunk2", "knowledge_name": "knowledge2"},
}
- vector_storge = VectorEmbeddingFactory.build_vector_storge('Pinecone', test_data)
+ vector_storge = VectorEmbeddingFactory.build_vector_storage('Pinecone', test_data)
mock_pinecone.assert_called_once_with(
[1,2], [[1,2,3],[4,5,6]], [{"text": "test", "chunk": "chunk", "knowledge_name": "knowledge"}, {"text": "test2", "chunk": "chunk2", "knowledge_name": "knowledge2"}]
)
- vector_storge = VectorEmbeddingFactory.build_vector_storge('Qdrant', test_data)
+ vector_storge = VectorEmbeddingFactory.build_vector_storage('Qdrant', test_data)
mock_qdrant.assert_called_once_with(
[1,2], [[1,2,3],[4,5,6]], [{"text": "test", "chunk": "chunk", "knowledge_name": "knowledge"}, {"text": "test2", "chunk": "chunk2", "knowledge_name": "knowledge2"}]