Merge pull request #88 from manubot/multiprovider-gh-var-docs

Adds provider, model vars docs to README

Author: falquaddoomi

.pre-commit-config.yaml

@@ -32,7 +32,7 @@ repos:
     hooks:
     -   id: poetry-check
 -   repo: https://github.com/tox-dev/pyproject-fmt
-    rev: "v2.5.1"
+    rev: "v2.6.0"
     hooks:
     -   id: pyproject-fmt
 -   repo: https://github.com/rhysd/actionlint

README.md

@@ -40,6 +40,25 @@ Start with a manuscript repo [forked from Manubot rootstock](https://github.com/
 1. If you haven't already, follow the directions above to create an account and get an API key for your chosen model provider.
 1. In your fork's "⚙️ Settings" tab, make a new Actions repository secret with the name `PROVIDER_API_KEY` and paste in your API key as the secret.
 
+If you prefer to select less options when running the workflow, you can optionally set up default values for the model provider and model at either the repo or organization level.
+
+In your fork's "⚙️ Settings" tab, you can optionally create the folllowing Actions repository variables:
+- `AI_EDITOR_MODEL_PROVIDER`: Either "openai" or "anthropic"; sets this as the default if "(repo default)" was selected in the workflow parameters.
+  If this is unspecified and "(repo default)" is selected, the workflow will throw an error.
+- `AI_EDITOR_LANGUAGE_MODEL`: For the given provider, what model to use if the "model" field in the workflow parameters was left empty.
+  If this is unspecified, Manubot AI Editor will select the default model for your chosen provider.
+
+### Multiple Providers
+
+In case you want to use several providers in the same repo, you'll have to register an API key for each provider you intend to use.
+Like `PROVIDER_API_KEY`, these keys are also registered as GitHub secrets, and can be specified at either the repository or organizational level.
+
+We currently support the following secrets, with more to follow as we integrate more providers:
+- `OPENAI_API_KEY`: the API key for the "openai" provider
+- `ANTHROPIC_API_KEY`: the API key for the "anthropic" provider
+
+See [the API key variables docs](https://github.com/manubot/manubot-ai-editor/blob/main/docs/env-vars.md#provider-api-key-configuration) for more information.
+
 ### Configuring prompts
 
 In order to revise your manuscript, prompts must be provided to the AI model.
@@ -140,8 +159,7 @@ specific encodings, you can specify the input encoding with the
 `AI_EDITOR_SRC_ENCODING` and the output encoding with the
 `AI_EDITOR_DST_ENCODING` environment variables.
 
-See
-[these variables' help docs](https://github.com/manubot/manubot-ai-editor/blob/main/docs/env-vars.md#encodings)
+See[these variables' help docs](https://github.com/manubot/manubot-ai-editor/blob/main/docs/env-vars.md#encodings)
 for more information.
 
 Also, see [Python 3 Docs: Standard Encodings](https://docs.python.org/3/library/codecs.html#standard-encodings) for

docs/env-vars.md

@@ -16,15 +16,11 @@ support are "openai" for OpenAI and "anthropic" for Anthropic.
 
 ## Provider API Key Configuration
 
-For providers that require API keys, you can specify an API key specific
-to that provider via an environment variable named `<PROVIDER>_API_KEY`.
-For example, for OpenAI, the API key variable would be named `OPENAI_API_KEY`
-and for Anthropic, it would be `ANTHROPIC_API_KEY`.
-
-Alternatively, you can use the environment variable `PROVIDER_API_KEY` to
-set an API key that will be used for all providers.
-If both a provider-specific key and `PROVIDER_API_KEY` are set, the provider-
-specific key will take precedence.
+For providers that require API keys, you can specify an API key specific to that provider via an environment variable named `<PROVIDER>_API_KEY`.
+For example, for OpenAI, the API key variable would be named `OPENAI_API_KEY` and for Anthropic, it would be `ANTHROPIC_API_KEY`.
+
+Alternatively, you can use the environment variable `PROVIDER_API_KEY` to set an API key that will be used for all providers.
+If both a provider-specific key and `PROVIDER_API_KEY` are set, the provider-specific key will take precedence.
 
 ## Model Configuration