LLM ask AI

llm-ask-ai

By Bernhard Enders

An Espanso package that enables users to quickly send prompts to a local (e.g. Ollama or LM Studio) or remote LLM API calls (OpenAI standard) and insert the AI-generated response directly into any text field.

LLM

artificial intelligence

prompt

Espanso

Ollama

OpenAI

Paste this command in a terminal to install the package

espanso install llm-ask-ai

Description

Source

llm-ask-ai

Requirements

Espanso installed and running
Python 3.9+ installed and available on your system path
Required Python packages:
openai
and
python-dotenv
(see
requirements.txt
)
Access to a local LLM (such as Ollama or LM Studio) or a remote OpenAI compatible API (in this case, you will need to provide your API key in the
.env
file)

Configuration

Edit the

.env

environment file inside the package directory () to set the

API_KEY

BASE_URL

and

MODEL

. Example:

API_KEY=ollama
BASE_URL=http://localhost:11434/v1
MODEL=llama3.2

NOTE: don't forget to pull the Ollama model first. In the case above, just issue the command:

ollama pull llama3.2

Usage

Type the Espanso trigger for this package (
:ask:ai
) in any text field.
Enter your desired AI prompt when asked.
The AI-generated response will be inserted automatically (this can take several seconds, so choose a small and low latency LLM model).

Example

Type:

:ask:ai

Then enter:

Summarize the following text: ...

Press the "Submit" button or "CTRL + Enter" to send the request. The response from your configured LLM will appear in place.

Troubleshooting

Make sure
python
(not
python3
) executable is available globally on your system's PATH environment variable and that the required packages are installed.
Certain Linux distributions, such as Debian, use only a
python3
executable and don't include a
python
executable or symlink. In this case, you may need to install the
python-is-python3
package (recommended), create a symbolic link, or run a command like this:
```
sudo update-alternatives --install /usr/bin/python python /usr/bin/python3 10
```
Make sure you have the necessary Python packages installed globally, either by running
pip install openai python-dotenv
or by using a command like
sudo apt install python3-openai python3-dotenv
.
Check that your BASE_URL endpoint, MODEL and API_KEY are correctly set in the
.env
file (located in the Espanso config directory:
%CONFIG%/match/packages/llm-ask-ai/.env
).
After sending the request, make sure the cursor doesn’t lose focus and remains blinking at the correct insertion point. If it doesn’t, just click to place it right after the trigger string like this :ask:ai|
Review Espanso logs for errors.

License

MIT

Author

Bernhard Enders

Links

package.yml

matches:
  - trigger: ":ask:ai"
    replace: "{{output}}"
    vars:
      - name: "prompt"
        type: "form"
        params:
          layout: "What would you like to ask the Artificial intelligence LLM today?\n\n[[text]]"
          fields:
            text:
              multiline: true
      - name: "output"
        type: "script"
        params:
          args:
            - python
            - "%CONFIG%/match/packages/llm-ask-ai/ask-ai.py"
            - "{{prompt.text}}"

Made with