fix: Improvements Jan 22 (#734)

Co-authored-by: Charles Packer <[email protected]>
Co-authored-by: dboyliao <[email protected]>
Co-authored-by: Shubham Naik <[email protected]>
Co-authored-by: Shubham Naik <[email protected]>
Co-authored-by: Caren Thomas <[email protected]>
Co-authored-by: Sarah Wooders <[email protected]>
Co-authored-by: Nuno Rocha <[email protected]>
Co-authored-by: Theo Conrads <[email protected]>
Co-authored-by: Jyotirmaya Mahanta <[email protected]>
Co-authored-by: Stephan Fitzpatrick <[email protected]>
Co-authored-by: Stephan Fitzpatrick <[email protected]>
Co-authored-by: mlong93 <[email protected]>
Co-authored-by: Mindy Long <[email protected]>
Co-authored-by: Krishnakumar R (KK) <[email protected]>

.github/workflows/docker-image.yml

@@ -8,7 +8,7 @@ on:
 jobs:
   build:
     runs-on: ubuntu-latest
-    
+
     steps:
     - name: Login to Docker Hub
       uses: docker/login-action@v3
@@ -17,7 +17,7 @@ jobs:
         password: ${{ secrets.DOCKERHUB_TOKEN }}
 
     - uses: actions/checkout@v3
-    
+
     - name: Set up QEMU
       uses: docker/setup-qemu-action@v3
 
@@ -38,4 +38,3 @@ jobs:
           letta/letta:latest
           memgpt/letta:${{ env.CURRENT_VERSION }}
           memgpt/letta:latest
-          

.github/workflows/letta-code-sync.yml

@@ -0,0 +1,19 @@
+name: Sync Code
+
+on:
+  push:
+    branches:
+      - main
+
+jobs:
+  notify:
+    runs-on: ubuntu-latest
+    if: ${{ !contains(github.event.head_commit.message, '[sync-skip]') }}
+    steps:
+      - name: Trigger repository_dispatch
+        run: |
+          curl -X POST \
+            -H "Authorization: token ${{ secrets.SYNC_PAT }}" \
+            -H "Accept: application/vnd.github.v3+json" \
+            https://api.github.com/repos/letta-ai/letta-cloud/dispatches \
+            -d '{"event_type":"oss-update"}'

CONTRIBUTING.md

@@ -19,18 +19,46 @@ Now, let's bring your new playground to your local machine.
 git clone https://github.com/your-username/letta.git
 ```
 
-### 🧩 Install Dependencies
+### 🧩 Install dependencies & configure environment
+
+#### Install poetry and dependencies
 
 First, install Poetry using [the official instructions here](https://python-poetry.org/docs/#installation).
 
-Once Poetry is installed, navigate to the Letta directory and install the Letta project with Poetry:
+Once Poetry is installed, navigate to the letta directory and install the Letta project with Poetry:
 ```shell
-cd Letta
+cd letta
 poetry shell
 poetry install --all-extras
 ```
+#### Setup PostgreSQL environment (optional)
+
+If you are planning to develop letta connected to PostgreSQL database, you need to take the following actions.
+If you are not planning to use PostgreSQL database, you can skip to the step which talks about [running letta](#running-letta-with-poetry).
+
+Assuming you have a running PostgreSQL instance, first you need to create the user, database and ensure the pgvector
+extension is ready. Here are sample steps for a case where user and database name is letta and assumes no password is set:
+
+```shell
+createuser letta
+createdb letta --owner=letta
+psql -d letta -c 'CREATE EXTENSION IF NOT EXISTS vector'
+```
+Setup the environment variable to tell letta code to contact PostgreSQL database:
+```shell
+export LETTA_PG_URI="postgresql://${POSTGRES_USER:-letta}:${POSTGRES_PASSWORD:-letta}@localhost:5432/${POSTGRES_DB:-letta}"
+```
+
+After this you need to prep the database with initial content. You can use alembic upgrade to populate the initial
+contents from template test data. Please ensure to activate poetry environment using `poetry shell`.
+```shell
+alembic upgrade head
+```
+
+#### Running letta with poetry
 
 Now when you want to use `letta`, make sure you first activate the `poetry` environment using poetry shell:
+
 ```shell
 $ poetry shell
 (pyletta-py3.12) $ letta run

README.md

@@ -9,7 +9,7 @@
 <div align="center">
 <h1>Letta (previously MemGPT)</h1>
 
-**☄️ New release: Letta Agent Development Environment (_read more [here](#-access-the-letta-ade-agent-development-environment)_) ☄️**
+**☄️ New release: Letta Agent Development Environment (_read more [here](#-access-the-ade-agent-development-environment)_) ☄️**
 
 <p align="center">
   <picture>
@@ -23,7 +23,7 @@
 
 <h3>
 
-[Homepage](https://letta.com) // [Documentation](https://docs.letta.com) // [ADE](https://app.letta.com) // [Letta Cloud](https://forms.letta.com/early-access)
+[Homepage](https://letta.com) // [Documentation](https://docs.letta.com) // [ADE](https://docs.letta.com/agent-development-environment) // [Letta Cloud](https://forms.letta.com/early-access)
 
 </h3>
 
@@ -80,12 +80,12 @@ docker run \
 
 Once the Letta server is running, you can access it via port `8283` (e.g. sending REST API requests to `http://localhost:8283/v1`). You can also connect your server to the Letta ADE to access and manage your agents in a web interface.
 
-### 👾 Access the [Letta ADE (Agent Development Environment)](https://app.letta.com)
+### 👾 Access the ADE (Agent Development Environment)
 
 > [!NOTE]
-> The Letta ADE is a graphical user interface for creating, deploying, interacting and observing with your Letta agents.
->
-> For example, if you're running a Letta server to power an end-user application (such as a customer support chatbot), you can use the ADE to test, debug, and observe the agents in your server. You can also use the ADE as a general chat interface to interact with your Letta agents.
+> For a guided tour of the ADE, watch our [ADE walkthrough on YouTube](https://www.youtube.com/watch?v=OzSCFR0Lp5s), or read our [blog post](https://www.letta.com/blog/introducing-the-agent-development-environment) and [developer docs](https://docs.letta.com/agent-development-environment).
+
+The Letta ADE is a graphical user interface for creating, deploying, interacting and observing with your Letta agents. For example, if you're running a Letta server to power an end-user application (such as a customer support chatbot), you can use the ADE to test, debug, and observe the agents in your server. You can also use the ADE as a general chat interface to interact with your Letta agents.
 
 <p align="center">
   <picture>

letta/__init__.py

@@ -1,4 +1,4 @@
-__version__ = "0.6.9"
+__version__ = "0.6.13"
 
 
 # import clients

letta/client/client.py

@@ -442,7 +442,8 @@ class RESTClient(AbstractClient):
     def __init__(
         self,
         base_url: str,
-        token: str,
+        token: Optional[str] = None,
+        password: Optional[str] = None,
         api_prefix: str = "v1",
         debug: bool = False,
         default_llm_config: Optional[LLMConfig] = None,
@@ -458,11 +459,18 @@ def __init__(
             default_llm_config (Optional[LLMConfig]): The default LLM configuration.
             default_embedding_config (Optional[EmbeddingConfig]): The default embedding configuration.
             headers (Optional[Dict]): The additional headers for the REST API.
+            token (Optional[str]): The token for the REST API when using managed letta service.
+            password (Optional[str]): The password for the REST API when using self hosted letta service.
         """
         super().__init__(debug=debug)
         self.base_url = base_url
         self.api_prefix = api_prefix
-        self.headers = {"accept": "application/json", "authorization": f"Bearer {token}"}
+        if token:
+            self.headers = {"accept": "application/json", "Authorization": f"Bearer {token}"}
+        elif password:
+            self.headers = {"accept": "application/json", "X-BARE-PASSWORD": f"password {password}"}
+        else:
+            self.headers = {"accept": "application/json"}
         if headers:
             self.headers.update(headers)
         self._default_llm_config = default_llm_config

letta/orm/job_usage_statistics.py

@@ -0,0 +1,30 @@
+from typing import TYPE_CHECKING, Optional
+
+from sqlalchemy import ForeignKey
+from sqlalchemy.orm import Mapped, mapped_column, relationship
+
+from letta.orm.sqlalchemy_base import SqlalchemyBase
+
+if TYPE_CHECKING:
+    from letta.orm.job import Job
+
+
+class JobUsageStatistics(SqlalchemyBase):
+    """Tracks usage statistics for jobs, with future support for per-step tracking."""
+
+    __tablename__ = "job_usage_statistics"
+
+    id: Mapped[int] = mapped_column(primary_key=True, doc="Unique identifier for the usage statistics entry")
+    job_id: Mapped[str] = mapped_column(
+        ForeignKey("jobs.id", ondelete="CASCADE"), nullable=False, doc="ID of the job these statistics belong to"
+    )
+    step_id: Mapped[Optional[str]] = mapped_column(
+        nullable=True, doc="ID of the specific step within the job (for future per-step tracking)"
+    )
+    completion_tokens: Mapped[int] = mapped_column(default=0, doc="Number of tokens generated by the agent")
+    prompt_tokens: Mapped[int] = mapped_column(default=0, doc="Number of tokens in the prompt")
+    total_tokens: Mapped[int] = mapped_column(default=0, doc="Total number of tokens processed by the agent")
+    step_count: Mapped[int] = mapped_column(default=0, doc="Number of steps taken by the agent")
+
+    # Relationship back to the job
+    job: Mapped["Job"] = relationship("Job", back_populates="usage_statistics")

letta/server/rest_api/routers/v1/agents.py

@@ -560,6 +560,9 @@ async def process_message_background(
         )
         server.job_manager.update_job_by_id(job_id=job_id, job_update=job_update, actor=actor)
 
+        # Add job usage statistics
+        server.job_manager.add_job_usage(job_id=job_id, usage=result.usage, actor=actor)
+
     except Exception as e:
         # Update job status to failed
         job_update = JobUpdate(

pyproject.toml

@@ -1,6 +1,7 @@
 [tool.poetry]
 name = "letta"
-version = "0.6.9"
+
+version = "0.6.13"
 packages = [
     {include = "letta"},
 ]
@@ -21,7 +22,7 @@ questionary = "^2.0.1"
 pytz = "^2023.3.post1"
 tqdm = "^4.66.1"
 black = {extras = ["jupyter"], version = "^24.2.0"}
-setuptools = "^68.2.2"
+setuptools = "^70"
 datasets = { version = "^2.14.6", optional = true}
 prettytable = "^3.9.0"
 pgvector = { version = "^0.2.3", optional = true }
@@ -47,7 +48,7 @@ qdrant-client = {version="^1.9.1", optional = true}
 python-box = "^7.1.1"
 sqlmodel = "^0.0.16"
 autoflake = {version = "^2.3.0", optional = true}
-python-multipart = "^0.0.9"
+python-multipart = "^0.0.19"
 sqlalchemy-utils = "^0.41.2"
 pytest-order = {version = "^1.2.0", optional = true}
 pytest-asyncio = {version = "^0.23.2", optional = true}
@@ -56,7 +57,7 @@ httpx-sse = "^0.4.0"
 isort = { version = "^5.13.2", optional = true }
 docker = {version = "^7.1.0", optional = true}
 nltk = "^3.8.1"
-jinja2 = "^3.1.4"
+jinja2 = "^3.1.5"
 locust = {version = "^2.31.5", optional = true}
 wikipedia = {version = "^1.4.0", optional = true}
 composio-langchain = "^0.6.15"
@@ -79,6 +80,7 @@ e2b-code-interpreter = {version = "^1.0.3", optional = true}
 anthropic = "^0.43.0"
 letta_client = "^0.1.16"
 
+
 [tool.poetry.extras]
 postgres = ["pgvector", "pg8000", "psycopg2-binary", "psycopg2"]
 dev = ["pytest", "pytest-asyncio", "pexpect", "black", "pre-commit", "datasets", "pyright", "pytest-order", "autoflake", "isort", "locust"]

tests/integration_test_tool_execution_sandbox.py

@@ -195,6 +195,14 @@ def composio_gmail_get_profile_tool(test_user):
     yield tool
 
 
+@pytest.fixture
+def composio_gmail_get_profile_tool(test_user):
+    tool_manager = ToolManager()
+    tool_create = ToolCreate.from_composio(action_name="GMAIL_GET_PROFILE")
+    tool = tool_manager.create_or_update_tool(pydantic_tool=Tool(**tool_create.model_dump()), actor=test_user)
+    yield tool
+
+
 @pytest.fixture
 def clear_core_memory_tool(test_user):
     def clear_memory(agent_state: "AgentState"):
@@ -418,6 +426,14 @@ def test_local_sandbox_e2e_composio_star_github_without_setting_db_env_vars(
     assert result.func_return["details"] == "Action executed successfully"
 
 
+@pytest.mark.local_sandbox
+def test_local_sandbox_e2e_composio_star_github_without_setting_db_env_vars(
+    mock_e2b_api_key_none, check_composio_key_set, composio_github_star_tool, test_user
+):
+    result = ToolExecutionSandbox(composio_github_star_tool.name, {"owner": "letta-ai", "repo": "letta"}, user=test_user).run()
+    assert result.func_return["details"] == "Action executed successfully"
+
+
 @pytest.mark.local_sandbox
 def test_local_sandbox_external_codebase(mock_e2b_api_key_none, custom_test_sandbox_config, external_codebase_tool, test_user):
     # Set the args

tests/test_tool_schema_parsing.py

@@ -203,4 +203,5 @@ def test_composio_tool_schema_generation(openai_model: str, structured_output: b
             print(f"Successfully called OpenAI using schema {schema} generated from {action_name}\n\n")
         except:
             print(f"Failed to call OpenAI using schema {schema} generated from {action_name}\n\n")
+
             raise