tulip_agent
The tulip agent is a powerful tool designed to leverage large language models (LLMs) with access to a wide range of tools. It aims to reduce costs associated with LLM usage by efficiently utilizing tools and overcoming limitations of API calls or context windows. Developers can customize the agent's behavior through instructions, overriding default prompts and tailoring it to specific tasks. The agent offers various configurations, from basic LLM agents to fully autonomous versions capable of dynamically managing their tool libraries. It's particularly useful for complex tasks requiring access to numerous tools and adaptable workflows. The agent's architecture includes function analysis, a tool library with semantic search, and diverse agent types to suit different needs. It provides a flexible and scalable solution for integrating LLMs with external tools.
The tulip agent solves the problem of efficiently managing and utilizing a large number of tools with LLMs, which can be costly and complex. Instead of manually chaining together multiple API calls or struggling with context window limitations, developers can use this agent to automate tool selection and execution, streamlining workflows and reducing operational expenses.
CAPABILITIES & CONSTRAINTS
USE CASES
README
# 🌷🤖 tulip agent
[](https://arxiv.org/abs/2407.21778)
[](https://opensource.org/license/bsd-3-clause)
[](https://github.com/psf/black)
A reference implementation for the `tulip agent`, an LLM-backed agent with access to a large number of tools via a tool library.
This approach reduces costs, enables the use of tool sets that exceed API limits or context windows, and increases flexibility with regard to the tool set used.
## Key components
🔬 **Function analysis** \
Generates OpenAI API compatible tool descriptions for Python functions via introspection
🌷 **Tool library** \
Combines a vector store for semantic search among tools and tool execution
🤖 **Agents**\
Specifying `instructions` for an agent completely overrides the base system prompts to avoid contradictions.
You can append custom instructions to the default prompts in `tulip_agent.agents.prompts`.
* Baseline, without tool library
* `BaseAgent`: LLM agent without tool access
* `NaiveToolAgent`: Includes tool descriptions for all tools available
* `CotToolAgent`: Extends the `NaiveToolAgent` with a planning step that decomposes the user input into subtasks
* Tulip variations with access to a tool library
* `MinimalTulipAgent`: Minimal implementation; searches for tools based on the user input directly
* `NaiveTulipAgent`: Naive implementation; searches for tools with a separate tool call
* `CotTulipAgent`: COT implementation; derives a plan for the necessary steps and searches for suitable tools
* `InformedCotTulipAgent`: Same as `CotTulipAgent`, but with a brief description of the tool library's contents
* `PrimedCotTulipAgent`: Same as `CotTulipAgent`, but primed with tool names based on an initial search with the user request
* `OneShotCotTulipAgent`: Same as `CotTulipAgent`, but the system prompt included a brief example
* `AutoTulipAgent`: Fully autonomous variant; can use the search tool at any time and modify its tool library with CRUD operations
* `DfsTulipAgent`: DFS inspired variant that leverages a DAG for keeping track of tasks and suitable tools, can create new tools
📊 **Evaluation**
* `math_eval`: Math evaluation
* `robo_eval`: Robotics evaluation using tools created for [AttentiveSupport](https://github.com/HRI-EU/AttentiveSupport)
📝 **Examples** \
See `./examples`
## Setup
* Make sure to set the environment variables required by the API of your choice. Currently supported:
* OpenAI: `OPENAI_API_KEY`, see the [official instructions](https://help.openai.com/en/articles/5112595-best-practices-for-api-key-safety)
* Azure: `AZURE_OPENAI_API_KEY`, `AZURE_API_VERSION`, and `AZURE_OPENAI_ENDPOINT`
* OpenAI compatible endpoints: `OAI_COMPATIBLE_BASE_URL` and `OAI_COMPATIBLE_API_KEY` for OpenAI compatible endpoints, such as Ollama
* Install with `uv venv --allow-existing && uv sync` or `pip install -e .`
* Check out the `examples`, the robot evaluation in `src/eval/robo_eval`, and `examples/local_examples.py` for a local setup
## Dev notes
* Python v3.10.11 recommended, higher versions may lead to issues with chroma during installation
* [Pre-commit hooks](https://pre-commit.com/) - install with `(uv run) pre-commit install`
* Linting: [ruff](https://github.com/astral-sh/ruff)
* Formatting: [black](https://github.com/psf/black)
* Import sorting: [isort](https://github.com/PyCQA/isort)
* Tests: Run with `(uv run) python -m unittest discover tests/`
## Known issues
### SQLite version incompatibility
See these [troubleshooting instructions](https://docs.trychroma.com/troubleshooting#sqlite)
1. On Linux install pysqlite3-binary: `uv add pysqlite3-binary`
2. Add the following to `lib/python3.10/site-packages/chromadb/__init__.py` in your venv
```python
__import__('pysqlite3')
import sys
sys.modules['sqlite3'] = sys.modules.pop('pysqlite3')
```
### Running the example results in a ModuleNotFoundError
Make sure to install the package itself, e.g., with `uv sync` or `pip install -e .` \
Then run the example with `uv run examples/calculator_example.py`
PUBLIC HISTORY
IDENTITY
Identity inferred from code signals. No PROVENANCE.yml found.
Is this yours? Claim it →METADATA
README BADGE
Add to your README:
