feat(agent): native function-calling for cloud providers - #359
Merged
Conversation
…ol use) Add provider.chatTools using the real OpenAI tools/tool_calls and Anthropic tool_use APIs — so big/complex args (SVG, code, multi-step plans) are encoded by the provider, not hand-formatted by the model as fragile JSON-in-prose. useAgentChat now runs a NATIVE tool loop for cloud providers (proper assistant/tool messages), falling back to the prompt-based JSON loop for on-device WebLLM. Executor running (file/param collection, chaining, param prompting) extracted into a shared runExecutor used by both loops.
Deploying with
|
| Status | Name | Latest Commit | Updated (UTC) |
|---|---|---|---|
View logs |
goodwebtools | d2ca939 | Aug 29 2026, 12:12 AM |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The reliability fix from v2 testing. Cloud providers now use their real tools API (OpenAI tool_calls / Anthropic tool_use) via provider.chatTools, so big args (SVG/code) and multi-step plans are encoded by the provider — not the fragile JSON-in-prose protocol that failed on GLM. useAgentChat runs a native tool loop for cloud, prompt-based loop for on-device WebLLM; shared runExecutor for file/param/chaining. Provider parsing unit-tested (openai+anthropic, mocked fetch); live loop is build+smoke.