shell-pilot

Installation

• For setup Ollama environment, Manual install instructions, ollama usage, and Ollama model library

curl -fsSL https://ollama.com/install.sh | sh

ollama pull llama2  # used llama2 by default

• For setup LocalAI environment, Manual, and LocalAI github

docker run -p 8080:8080 --name local-ai -ti localai/localai:latest-aio-cpu
# Do you have a Nvidia GPUs? Use this instead
# CUDA 11
# docker run -p 8080:8080 --gpus all --name local-ai -ti localai/localai:latest-aio-gpu-nvidia-cuda-11
# CUDA 12
# docker run -p 8080:8080 --gpus all --name local-ai -ti localai/localai:latest-aio-gpu-nvidia-cuda-12

• To install, run this in your terminal and provide your OpenAI API key when asked.

curl -sS -o spilot_install.sh https://raw.githubusercontent.com/reid41/shell-pilot/main/spilot_install.sh
bash spilot_install.sh

• Set your local Ollama server ip in configuration file spilot_common.sh if not set during the installation

OLLAMA_SERVER_IP=<ollama server ip address>

• You can also set the other parameters in spilot_common.sh before using.

e.g.
TEMPERATURE=0.6
MAX_TOKENS=4096
MODEL_OPENAI=gpt-3.5-turbo
MODEL_OLLAMA=llama2
CONTEXT=false
MULTI_LINE_PROMPT=false
ENABLE_DANGER_FLAG=false

Manual Installation

If you want to install it manually, all you have to do is:

• Download the shell-pilot project files in

git clone https://github.com/reid41/shell-pilot.git

cd shell-pilot/

• If you want to reset the script path(/usr/local/bin/ by default) or output store path(~/spilot_files_dir by default), try below:

# define the config dir if need to reset
new_config_path="/new/path/to/config/"
# define the tmp or output files dir
new_files_dir="/new/path/to/files/"

# create it if new
[[ ! -d ${new_config_path} ]] && mkdir ${new_config_path} -p
[[ ! -d "${new_config_path}/plugins" ]] && mkdir ${new_config_path}/plugins -p
[[ ! -d ${new_files_dir} ]] && mkdir ${new_files_dir} -p

# reset it
sed -i "s|SHELL_PILOT_CONFIG_PATH=\"/usr/local/bin/\"|SHELL_PILOT_CONFIG_PATH=\"$new_config_path\"|" s-pilot
sed -i "s|SPILOT_FILES_DEFAULT_DIR=~/spilot_files_dir|SPILOT_FILES_DEFAULT_DIR=$new_files_dir|" spilot_common.sh

# add ollama server host
ollama_server_ip_address=<ip>
echo "OLLAMA_SERVER_IP=${ollama_server_ip_address}" >> spilot_common.sh

# add localai server host
localai_server_ip_address=<ip>
echo "LOCALAI_SERVER_IP=${localai_server_ip_address}" >> spilot_common.sh

• set the permissions

chmod +x s-pilot spilot_common.sh spilot_llm_rq_apis.sh plugins/*.sh

• Move the files to the dir

cp s-pilot spilot_common.sh spilot_llm_rq_apis.sh ${new_config_path}

cp plugins/*.sh ${new_config_path}/plugins

• Add settings into the profile file

# profile, e.g. .bash_profile
the_profile_file=$HOME/.bash_profile

# add the script/config path
echo "export PATH\=\$PATH\:${new_config_path}" >> $the_profile_file

# add source alias for alias option
echo "alias ss-pilot='source s-pilot'" >> $the_profile_file

# openai/mistral key if need
# openai
openai_api_key_value=<key>
echo "export OPENAI_KEY=${openai_api_key_value}" >> $the_profile_file

# mistral ai
mistralai_api_key_value=<key>
echo "export MISTRAL_API_KEY=${mistralai_api_key_value}" >> $the_profile_file

# zhipuai ai
zhipuai_api_key_value=<key>
echo "export ZHIPUAI_API_KEY=${zhipuai_api_key_value}" >> $the_profile_file

# anthropic ai
anthropic_api_key_value=<key>
echo "export ANTHROPIC_API_KEY=${anthropic_api_key_value}" >> $the_profile_file

# moonshot ai
moonsho