Requirements

• Python 3.10+
• An API key for at least one LLM provider (or a local endpoint)

Install cliver-core

The project is on pre-releases, so pin the exact version — an exact pin selects the pre-release without --pre:

pip install cliver-core==0.3.0a5

The core package includes the MCP client (stdio and streamable HTTP).

Optional local models

In-process embedding and rerank backends pull torch, so they are opt-in:

pip install "cliver-core[local-st]==0.3.0a5"     # sentence-transformers embeddings
pip install "cliver-core[local-flag]==0.3.0a5"   # FlagEmbedding embeddings + rerank
pip install "cliver-core[local]==0.3.0a5"        # both

Verify Installation

python -c "from cliver_core import AgentCore; print(AgentCore)"

Work on the workspace

The repository is a uv workspace with two packages: cliver-core and the cliver CLI application.

git clone https://github.com/cliver-project/CLIver.git
cd CLIver

make init          # venv + workspace dependencies
make init-local    # + optional local model backends
make test          # run both packages' tests

Next Steps

Configure model bindings to connect cliver-core to your LLM of choice.