Pull Request Explorer
| Title | Author | Size | AI | Cycle Time | Review | Merged |
|---|---|---|---|---|---|---|
| fix(docs): resolve model placeholders in code examples | gr2m | L | No | 89.8h | - | Sep 29, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| Reuse prepared image options in image components | gnoff | L | No | 12.8h | - | Sep 29, 2026 |
SummaryHealth AssessmentAI Details
Confidence:
0.95
Tech Stack
Languages:
Typescript, Javascript
Frameworks:
Nextjs
|
||||||
| Scope client component loading metrics to HTML renders | gnoff | XL | No | 44.4h | 1.5h | Sep 29, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| docs: fix docsgraph page defects and resolve placeholders in Markdown routes | molebox | XL | No | 6.9h | 6.9h | Sep 29, 2026 |
SummaryDocsFix docsgraph defects and placeholder resolution Corrects broken documentation samples and removes placeholder tokens, ensuring accurate code examples for developers. Health Assessment
Large
Low
Low
AI DetailsTech Stack
Languages:
Typescript
Frameworks:
Nextjs, React
|
||||||
| fix: Claude Code streams omitted tool progress and individual response completion signals. | ai-sdk-factory | M | AI | 3.6h | 0.4h | Sep 29, 2026 |
SummaryBug FixFix Claude Code stream progress signals Ensures Claude Code streams correctly emit progress and completion events, improving reliability of harness integration. Health Assessment
Medium
Low
Medium
AI Details
Category:
Code AI
Confidence:
0.80
Tech Stack
Languages:
Typescript
|
||||||
| fix: preserve HTTP status codes during prerendering | jamiboym | L | No | 153.4h | 59.7h | Sep 29, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| test: enable deploy coverage for five verified suites | jamiboym | S | No | 111.3h | 0.1h | Sep 29, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| fix: Sub-agent tool activity, text, progress, and background-task notifications were unavailable to Harness consumers for rendering and correlation. | ai-sdk-factory | L | AI | 0.7h | 0.3h | Sep 29, 2026 |
SummaryBug FixFix missing sub-agent notifications in Harness Ensures sub-agent and task notifications are correctly forwarded to Harness consumers, improving rendering and correlation. Health Assessment
Large
Low
Low
AI Details
Category:
Code AI
Confidence:
0.95
Tech Stack
Languages:
Typescript
|
||||||
| Version Packages | github-actions | L | AI | 0.9h | 0.7h | Sep 29, 2026 |
SummaryChoreAutomated Release of SDK Packages Automated bumping of package versions and updating changelogs for multiple SDKs, enabling a smooth npm publish workflow. Health Assessment
Large
Low
Low
AI Details
Category:
Code AI
Confidence:
0.95
Tech Stack |
||||||
| fix(harness): expose `resolveSandboxCredentialEnvironment()` helper in favor of `createSandboxCredentialEnvironment()` and use it across bridge backed adapters | felixarntz | XL | AI | 21.8h | - | Sep 29, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| fix: workflow harness suppressing detach failures and returning stale resume state | ai-sdk-factory | M | No | 25.2h | 25.2h | Sep 29, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| fix(openai): rewrite recursive Zod schemas that use `allOf` wrappers | n1ckoates | L | No | 9.9h | 9.9h | Sep 29, 2026 |
SummaryBug FixRewrite recursive Zod schemas for OpenAI compatibility This change rewrites recursive Zod schemas to avoid unsupported `allOf` wrappers, allowing OpenAI structured outputs to accept them. It reduces errors and improves reliability for developers using recursive data structures. Health Assessment
Large
Low
Low
AI Details
Confidence:
0.20
Tech Stack
Languages:
Typescript
|
||||||
| fix: Start/end callbacks and fast tool results were delayed behind slower tools in the same step. | ai-sdk-factory | L | No | 0.6h | 0.6h | Sep 29, 2026 |
SummaryBug FixFix delayed callbacks and tool results in harness Ensures tool callbacks and results are delivered in real-time, improving responsiveness and user experience. Health Assessment
Medium
Low
Low
AI Details
Confidence:
0.80
Tech Stack
Languages:
Typescript
|
||||||
| Document agent feedback field limits | aurorascharff | S | No | 0.4h | 0.1h | Sep 29, 2026 |
SummaryChoreDocument agent feedback field limits Adds documentation and tests for field-length limits in the agent feedback form, enabling agents to prepare valid report links and avoid generic decode failures. Health Assessment
Small
Low
Low
AI Details
Usage:
AI Reviewed
Category:
Code AI
Tools:
Copilot
Confidence:
0.95
Tech Stack
Languages:
Typescript
|
||||||
| fix(docs): add canonical URLs to resource landing pages | gr2m | XS | No | 76.7h | - | Sep 29, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| Turbopack: Don't force non-recursive watcher for additional roots | bgw | XS | No | 1.0h | 0.0h | Sep 29, 2026 |
SummaryBug FixTurbopack: Disable non-recursive watcher for roots Improves Turbopack compiler stability on macOS by avoiding slow non-recursive file watchers, preventing hangs for projects with many dependencies. Health Assessment
Small
Low
Low
AI DetailsTech Stack
Languages:
Rust
|
||||||
| [test] Stabilize middleware revalidation test | unstubbable | XS | No | 3.1h | 3.1h | Sep 28, 2026 |
SummaryTestStabilize middleware revalidation test Ensures consistent middleware revalidation, reducing test flakiness and improving reliability of page rendering. Health Assessment
Small
Low
Low
AI Details
Confidence:
0.10
Tech Stack
Languages:
Typescript
Frameworks:
Nextjs, React
|
||||||
| docs: add email as a security reporting option | caleb-vercel | XS | No | 0.2h | 0.2h | Sep 28, 2026 |
SummaryDocsAdd email as security reporting option Reporters can now email [email protected] for security issues. Health Assessment
Small
Low
Low
AI Details
Usage:
Authored by AI
Category:
Code AI
Tools:
Claude
Confidence:
0.95
Tech Stack |
||||||
| docs: point security reporting at the public Vercel HackerOne program | caleb-vercel | XS | No | 0.3h | 0.3h | Sep 28, 2026 |
SummaryDocspoint security reporting at public Vercel HackerOne program Adds reference to public security reporting program, improving transparency and encouraging responsible disclosure. Health Assessment
Small
Low
Low
AI Details
Usage:
Authored by AI
Category:
Code AI
Tools:
Claude
Confidence:
0.95
Tech Stack |
||||||
| test(ai): cover persisted aborted tool history | gr2m | M | No | 20.5h | - | Sep 28, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| docs(ai): document provider tool error exception in v6 | gr2m | M | No | 1.1h | - | Sep 28, 2026 |
SummaryChoreDocument provider tool error exception in v6 Clarifies provider tool error handling in v6, preserving error codes for downstream model requests and improving reliability for developers. Health Assessment
Small
Low
Low
AI Details
Confidence:
0.95
Tech Stack
Languages:
Typescript, Yaml
|
||||||
| Version Packages | github-actions | L | AI | 0.9h | 0.6h | Sep 28, 2026 |
SummaryChoreVersion Packages Automated release of package versions, updating changelogs and package.json files for multiple SDK packages, ensuring consumers receive the latest stable releases. Health Assessment
Medium
Low
Low
AI Details
Category:
Code AI
Tech Stack |
||||||
| docs: fix documented typos | gr2m | S | No | 44.8h | 32.1h | Sep 28, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| test: execute packed SDK consumer bundles with Zod 3 and 4 | gr2m | XL | No | 1.1h | - | Sep 28, 2026 |
SummaryTestAdd packed SDK consumer bundle tests with Zod Ensures SDK bundles execute correctly across Zod 3 and 4, preventing regressions in consumer builds and improving reliability of production Next.js bundles. Health Assessment
X-Large
Low
Low
AI Details
Confidence:
0.80
Tech Stack
Languages:
Typescript, Javascript, Yaml
Frameworks:
Nextjs
|
||||||
| test: scope loader build dependency update to Turbopack | jimmyhmiller | S | No | 0.7h | 0.7h | Sep 28, 2026 |
SummaryTestUpdate test for Turbopack build dependency assertion Adjust test to only run with Turbopack and replace fixed delay with polling to avoid CI failures. Health Assessment
Small
Low
Low
AI DetailsTech Stack
Languages:
Typescript
|
||||||
| chore(provider/gateway): update gateway model settings files | github-actions | XS | No | 4.9h | 4.6h | Sep 28, 2026 |
SummaryChoreUpdate gateway model settings files Automated update of gateway model settings files to surface newly released gateway models as IDE autocomplete suggestions. Health Assessment
Small
Low
Low
AI DetailsTech Stack
Languages:
Typescript
|
||||||
| chore(provider/gateway): update gateway model settings files | github-actions | XS | No | 124.9h | 124.6h | Sep 28, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| [v5.0] fix: prevent unexpected chat resumption after terminal text follows a completed tool call | ai-sdk-factory | M | No | 0.5h | 0.1h | Sep 28, 2026 |
SummaryBug FixFix chat resumption after terminal text Prevents the UI from auto‑submitting a new request when terminal error text follows a completed tool call, improving chat reliability. Health Assessment
Medium
Low
Low
AI Details
Confidence:
0.95
Tech Stack
Languages:
Typescript
|
||||||
| Backport: chore(provider/gateway): update gateway model settings files v6 | github-actions | XS | No | 4.8h | 4.6h | Sep 28, 2026 |
SummaryChoreUpdate gateway model settings for new models Automated update of gateway model settings to surface newly released models as IDE autocomplete suggestions. Health Assessment
Small
Low
Low
AI Details
Confidence:
0.95
Tech Stack
Languages:
Typescript
|
||||||
| Backport: chore(provider/gateway): update gateway model settings files v5 | github-actions | XS | No | 4.8h | 4.5h | Sep 28, 2026 |
SummaryChoreBackport: update gateway model settings Automated update of gateway model settings to surface new models in IDE autocomplete. Health Assessment
Small
Low
Low
AI DetailsTech Stack
Languages:
Typescript
|
||||||
| fix(ai,mcp): backport prototype-safe maps to v6 | gr2m | L | No | 0.4h | - | Sep 28, 2026 |
SummaryBug FixBackport prototype-safe maps to v6 Fixes prototype pollution in provider registry and MCP client, ensuring metadata integrity and preventing crashes. Health Assessment
Medium
Medium
Low
AI Details
Confidence:
0.75
Tech Stack
Languages:
Typescript, Javascript
|
||||||
| fix(ai): backport reconnect chat ID encoding to v5 | gr2m | M | No | 0.3h | - | Sep 28, 2026 |
SummaryBug FixBackport reconnect chat ID encoding to v5 Fixes URL encoding for reconnect chat IDs, ensuring correct routing and security across API base variations. Health Assessment
Medium
Low
Low
AI Details
Confidence:
0.95
Tech Stack
Languages:
Typescript
|
||||||
| fix(ai): backport reconnect chat ID encoding to v6 | gr2m | M | No | 0.3h | - | Sep 28, 2026 |
SummaryBug FixBackport reconnect chat ID encoding to v6 Ensures reconnect URLs correctly encode chat IDs, preventing path traversal and query misinterpretation, and adds comprehensive tests and documentation. Health Assessment
Medium
Low
Low
AI Details
Confidence:
0.75
Tech Stack
Languages:
Typescript
|
||||||
| [v6.0] fix: prevent unexpected chat resumption after terminal text follows a completed tool call | ai-sdk-factory | M | No | 0.3h | 0.1h | Sep 28, 2026 |
SummaryBug FixFix chat resumption after terminal text This fix stops the UI from mistakenly resuming chat after a tool call completes and terminal error text appears, ensuring accurate completion handling. Health Assessment
Medium
Low
Low
AI Details
Confidence:
0.75
Tech Stack
Languages:
Typescript
|
||||||
| fix(ai,mcp): backport prototype-safe maps to v5 | gr2m | L | No | 0.3h | - | Sep 28, 2026 |
SummaryBug FixBackport prototype-safe maps to v5 Fixes prototype-related issues in v5, ensuring safe map handling for provider metadata and MCP tools, improving stability and preventing crashes. Health Assessment
Medium
Low
Low
AI DetailsTech Stack
Languages:
Typescript
|
||||||
| Version Packages | github-actions | M | No | 0.5h | 0.3h | Sep 28, 2026 |
SummaryChoreVersion Packages Automated package version bump and release preparation for Vercel AI SDKs, ensuring consistent versioning across SDKs. Health Assessment
Medium
Low
Low
AI DetailsTech Stack |
||||||
| Version Packages | github-actions | L | No | 1.3h | 1.1h | Sep 28, 2026 |
SummaryChoreAutomated package version release Publishes updated package versions to npm, streamlining the release workflow. Health Assessment
Large
Low
Low
AI DetailsTech Stack |
||||||
| Version Packages | github-actions | XL | AI | 1.7h | 1.5h | Sep 28, 2026 |
SummaryChoreAutomated release of SDK packages Updates changelogs and package versions for multiple SDKs, enabling npm publication. Health Assessment
X-Large
Low
Low
AI Details
Usage:
Authored by AI
Category:
Code AI
Tools:
Claude
Confidence:
0.95
Tech Stack
Languages:
Javascript
|
||||||
| docs(ai): explain provider-executed tool error masking exception | gr2m | S | No | 0.4h | - | Sep 28, 2026 |
SummaryDocsDocument provider-executed tool error masking exception Clarifies how provider-executed tool errors are handled, ensuring error codes are preserved for downstream model requests and improving reliability of the AI SDK. Health Assessment
Small
Low
Low
AI DetailsTech Stack
Languages:
Typescript
|
||||||
| fix(ai): encode chat IDs in stream reconnect URLs | gr2m | M | No | 0.4h | - | Sep 28, 2026 |
SummaryBug FixEncode chat IDs in reconnect URLs Ensures chat IDs are safely encoded in reconnect URLs, preventing path traversal and query manipulation. This improves reliability and security of stream reconnections in the AI SDK. Health Assessment
Medium
Low
Low
AI Details
Confidence:
0.75
Tech Stack
Languages:
Typescript
|
||||||
| fix(ai,mcp): preserve prototype-named tools and provider metadata | gr2m | L | No | 0.5h | - | Sep 28, 2026 |
SummaryBug FixFix prototype-named tools and provider metadata Prevents crashes and incorrect provider resolution caused by prototype-named keys, ensuring reliable AI tool and provider metadata handling. This improves stability for applications using the AI SDK and MCP client. Health Assessment
Large
Low
Low
AI DetailsTech Stack
Languages:
Typescript, Javascript
|
||||||
| Remove CLAUDE.md symlinks in favor of AGENTS.md | jarrensj | XS | AI | 238.5h | 237.9h | Sep 28, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| Align ExecuteContext task accessor names | lukesandberg | S | No | 72.6h | 0.0h | Sep 28, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| [test] Stabilize responsive dev overlay header layout test | jimmyhmiller | S | No | 1.3h | 0.0h | Sep 28, 2026 |
SummaryTestStabilize responsive dev overlay header layout test Ensures UI overlay header layout remains stable across responsive changes, preventing future test failures. Health Assessment
Small
Low
Low
AI DetailsTech Stack
Languages:
Typescript
Frameworks:
Nextjs
|
||||||
| fix: resume hydrated partial static tool calls without losing streaming state | ai-sdk-factory | L | No | 93.7h | 74.7h | Sep 28, 2026 |
|
LLM analysis pending This PR has not been analyzed yet. |
||||||
| [v5.0] fix: preserve smoothStream provider metadata without loss or cross-delta leakage | ai-sdk-factory | L | No | 0.2h | 0.2h | Sep 28, 2026 |
SummaryBug FixFix smoothStream provider metadata preservation Ensures smoothStream retains provider metadata across chunks, preventing data loss and ensuring accurate content rendering for downstream consumers. Health Assessment
Medium
Low
Low
AI Details
Confidence:
0.80
Tech Stack
Languages:
Typescript
|
||||||
| [v6.0] fix: preserve smoothStream provider metadata without loss or cross-delta leakage | ai-sdk-factory | L | No | 0.2h | 0.2h | Sep 28, 2026 |
SummaryBug FixFix smoothStream provider metadata preservation Ensures smoothStream retains provider metadata across deltas, preventing data loss and cross-delta contamination, improving reliability of AI streaming responses. Health Assessment
Medium
Low
Low
AI Details
Confidence:
0.30
Tech Stack
Languages:
Typescript
|
||||||
| [v6.0] fix: preserve tool metadata from tool output chunks in UI message streams | ai-sdk-factory | M | No | 1.7h | 1.7h | Sep 28, 2026 |
SummaryBug FixPreserve tool metadata in UI message streams Ensures UI displays complete tool execution metadata, improving user visibility and debugging. Health Assessment
Medium
Low
Low
AI Details
Confidence:
0.30
Tech Stack
Languages:
Typescript
|
||||||
| fix(azure): reject resource names that would rewrite the request host | kkdawkins | M | No | 0.3h | 0.2h | Sep 28, 2026 |
SummaryBug FixReject unsafe Azure resource names This change rejects Azure resource names that could rewrite request hosts, preventing potential security risks. It ensures only valid DNS labels are accepted, safeguarding user-supplied configurations. Health Assessment
Medium
Low
Low
AI DetailsTech Stack
Languages:
Typescript
|
||||||
| docs(google): update schema limitations | n1ckoates | S | No | 0.3h | 0.0h | Sep 28, 2026 |
SummaryDocsUpdate Google schema limitation documentation This PR clarifies how the AI SDK handles complex Google schemas that trigger vague errors, guiding users to resolve them. It also removes outdated OpenAPI references from the Google provider docs. Health Assessment
Small
Low
Low
AI DetailsTech Stack |
||||||