Support Model

[nathro]

Add optional model selection for PR description generation

• Add a new optional model input to the action. This value is passed through to augmentcode/augment-agent to choose the LLM used for generation (e.g., gpt-4o, claude-3-5-sonnet). When omitted, the agent’s default model is used.
• Document the model input in the README inputs table with examples.
• Update the augmentcode/augment-agent reference from a pinned commit to the @v0 tag to track the latest minor updates of the agent.

Impact

• No breaking changes; existing workflows continue to work without modification.
• Provides teams control over quality/cost/provider by selecting a model.
• If strict reproducibility is needed, consider pinning a specific agent commit.

---

🤖 This description was generated automatically. Please react with 👍 if it's helpful or 👎 if it needs improvement.

[github-actions]

Thanks for adding the configurable model input and updating the action reference. I left two small, high-confidence suggestions focused on security and docs clarity.

[github-actions]

Using a mutable tag (@v0) makes the workflow susceptible to supply‑chain issues if the tag is moved. Consider pinning to an immutable commit SHA and keeping a comment for readability, for example:

uses: augmentcode/augment-agent@<commit_sha> # v0

This preserves security while still signaling the intended major/minor version.

[github-actions]

Nice addition. To improve discoverability, consider adding a quick usage example that shows how to set model in the workflow, e.g. in one of the examples:

with:
  # ...existing inputs
  model: gpt-4o

This helps users understand exactly where/how to pass the new input.

[mtpauly]

The example should probably be a model that's publicly available (sonnet4 or gpt5)

[nathro]

Yup, I forgot to check this. Will update.