Repository navigation
feat: refresh models and support AI SDK 7 - #55
Merged
Merged
Conversation
Contributor
Author
How to use the Graphite Merge QueueAdd either label to this PR to merge it via the merge queue:
You must have a Graphite account in order to use the merge queue. Sign up using this link. An organization admin has enabled the Graphite Merge Queue in this repository. Please do not merge from GitHub as this will restart CI on PRs being processed by the merge queue. This stack of pull requests is managed by Graphite. Learn more about stacking. |
anandpant
marked this pull request as ready for review
August 7, 2026 02:09
|
Bugbot is not enabled for your account, so this pull request was not reviewed. Enable Bugbot in the Cursor dashboard to get automatic reviews on future PRs. |
anandpant
force-pushed
the
feat/model-catalog-ai-sdk-7
branch
from
August 7, 2026 17:39
d9a591c to
632a498
Compare
Contributor
Author
Merge activity
|
## Summary - add enrollment-verified GPT-5.6, Claude 5, and Gemini 3.5/3.6 aliases and metadata - derive adapter specification versions from the installed AI SDK instead of hard-coding v3 - support AI SDK 7 stable and verified late v7.0 beta floors through peer dependency ranges - adapt Anthropic requests to the Foundry proxy's supported tool streaming and structured-output shapes - test packed consumer installs against current stable v7 and the supported beta floors in CI ## Validation - `pnpm run verify` (52 unit tests plus lint, typecheck, build, package audit, skill validation, and release policy) - packed consumer typecheck and runtime checks with current AI SDK 7/provider v4 releases - packed consumer typecheck and runtime checks with the declared beta floors - `actionlint .github/workflows/ci.yml` - `codex review --base main` plus incremental reviews of follow-up fixes; no open findings ## Live verification - verified the seven new aliases and their metadata against the active Foundry enrollment - ran the targeted Foundry capability harness for every new alias: 68 supported capability checks passed and 37 unsupported modality checks skipped as expected - live-tested the packed package with both the declared AI SDK 7 beta floors and current stable AI SDK 7/provider v4 releases - covered OpenAI, Anthropic, and Google text generation plus Anthropic tool calls and streaming structured output Live testing exposed Foundry incompatibilities with Anthropic eager tool streaming and native structured output. The adapter now normalizes those requests to the proxy-supported forms and verifies Claude 5 summarized reasoning output.
graphite-app
Bot
force-pushed
the
feat/model-catalog-ai-sdk-7
branch
from
August 7, 2026 17:42
632a498 to
b4b68aa
Compare
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
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.
Summary
Validation
pnpm run verify(52 unit tests plus lint, typecheck, build, package audit, skill validation, and release policy)actionlint .github/workflows/ci.ymlcodex review --base mainplus incremental reviews of follow-up fixes; no open findingsLive verification
Live testing exposed Foundry incompatibilities with Anthropic eager tool streaming and native structured output. The adapter now normalizes those requests to the proxy-supported forms and verifies Claude 5 summarized reasoning output.