Skip to content

test(W-18772941): test against LLM Gateway #60

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 34 commits into
base: main
Choose a base branch
from

Conversation

mdonnalley
Copy link
Contributor

@mdonnalley mdonnalley commented Jun 16, 2025

What does this PR do?

Adds a script (yarn test:llmg) to test that tools are successfully invoked across multiple LLMs

Developers will need to set SF_LLMG_API_KEY to the api key, which will be stored in 1password

Developers will also need a llmg-test.yml (which is gitignored) to tell the test which models and prompts to use. For example,

models:
  - llmgateway__OpenAIGPT35Turbo_01_25
  - llmgateway__OpenAIGPT4OmniMini
  - llmgateway__BedrockAnthropicClaude4Sonnet

prompts:
  - What's my salesforce username?
  - List all my orgs
  # send multiple messages to the LLM
  - - I am a Salesforce developer. My current working directory is ~/my-project. My current org is [email protected].
      - Deploy my project

At this point the script is meant to be used during development to help write tool descriptions and parameters that are effective across multiple LLMs.

In the future we could possibly leverage this to write regression tests (although that would be expensive!)

Screenshot 2025-06-18 at 2 21 21 PM

What issues does this PR fix or reference?

@W-18772941@

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants