feat(evals): add Custom Token Exchange × Next.js two-persona eval#57

Draft

brth31 wants to merge 23 commits into

mainfrom

feat/add-custom-token-exchange-nextjs-eval

Commits on Jun 22, 2026

feat(evals): add Custom Token Exchange × Next.js two-persona eval
brth31
committed
chore(deps): bump @google/gemini-cli from 0.45.2 to 0.46.0

dependabot[bot]
authored and
brth31
committed
chore(deps): bump braintrust from 3.10.0 to 3.17.0

dependabot[bot]
authored and
brth31
committed
fix(core): don't apply Bedrock model map to judge and recommendations (#33)

frederikprijck
authored and
brth31
committed
fix(swift-eval): accept fluent webAuth().start() chain in L5 grader (#34)

sanchitmehtagit
authored and
brth31
committed
chore(deps-dev): bump typescript-eslint from 8.61.0 to 8.61.1 (#40)

dependabot[bot]
authored and
brth31
committed
chore(deps-dev): bump @vitest/coverage-v8 from 4.1.8 to 4.1.9 (#41)

dependabot[bot]
authored and
brth31
committed
chore(deps-dev): bump eslint from 10.4.1 to 10.5.0 (#39)

dependabot[bot]
authored and
brth31
committed
chore(deps): bump @ai-sdk/openai from 3.0.69 to 3.0.71 (#42)

dependabot[bot]
authored and
brth31
committed
fix: strengthen express-api graders: verify issuer/audience via .env event, not source substrings (#45)

sanchitmehtagit
authored and
brth31
committed
chore(deps): bump ai from 6.0.201 to 6.0.206 (#44)

dependabot[bot]
authored and
brth31
committed
chore(deps): bump the openai-codex group across 1 directory with 2 updates (#36)

dependabot[bot]
authored and
brth31
committed
fix(gemini-cli): normalize mcp_ tool names to the mcp__ convention (#46)

frederikprijck
authored and
brth31
committed
chore: group Anthropic claude-code and claude-agent-sdk dependabot updates (#48)

frederikprijck
authored and
brth31
committed
chore(deps): bump the anthropic group with 2 updates (#49)

dependabot[bot]
authored and
brth31
committed
docs: Add Architecture.md file (#35)

sanchitmehtagit
authored and
brth31
committed
fix(copilot): bump @github/copilot-sdk to 1.0.1 and route through LLM proxy (#50)

frederikprijck
authored and
brth31
committed
feat: inject compile_command guidance into agent context files (#18)

frederikprijck
authored and
brth31
committed
docs(agents): require ARCHITECTURE.md diagram updates on drift (#52)

sanchitmehtagit
authored and
brth31
committed
feat: post-run compiles() grader (#51)

frederikprijck
authored and
brth31
committed
feat(mfa/react): add compile-time check to React MFA eval (#53)

sanchitmehtagit
authored and
brth31
committed
fix(compile): auto-run npm install before compile_command (#55)

sanchitmehtagit
authored and
brth31
committed
chore(evals): add compile_command and compiles() grader to CTE evals
brth31
committed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

feat(evals): add Custom Token Exchange × Next.js two-persona eval#57

feat(evals): add Custom Token Exchange × Next.js two-persona eval#57
brth31 wants to merge 23 commits into
mainfrom
feat/add-custom-token-exchange-nextjs-eval

Commits on Jun 22, 2026

feat(evals): add Custom Token Exchange × Next.js two-persona eval

chore(deps): bump @google/gemini-cli from 0.45.2 to 0.46.0

chore(deps): bump braintrust from 3.10.0 to 3.17.0

fix(core): don't apply Bedrock model map to judge and recommendations (#33)

fix(swift-eval): accept fluent webAuth().start() chain in L5 grader (#34)

chore(deps-dev): bump typescript-eslint from 8.61.0 to 8.61.1 (#40)

chore(deps-dev): bump @vitest/coverage-v8 from 4.1.8 to 4.1.9 (#41)

chore(deps-dev): bump eslint from 10.4.1 to 10.5.0 (#39)

chore(deps): bump @ai-sdk/openai from 3.0.69 to 3.0.71 (#42)

fix: strengthen express-api graders: verify issuer/audience via .env event, not source substrings (#45)

chore(deps): bump ai from 6.0.201 to 6.0.206 (#44)

chore(deps): bump the openai-codex group across 1 directory with 2 updates (#36)

fix(gemini-cli): normalize mcp_ tool names to the mcp__ convention (#46)

chore: group Anthropic claude-code and claude-agent-sdk dependabot updates (#48)

chore(deps): bump the anthropic group with 2 updates (#49)

docs: Add Architecture.md file (#35)

fix(copilot): bump @github/copilot-sdk to 1.0.1 and route through LLM proxy (#50)

feat: inject compile_command guidance into agent context files (#18)

docs(agents): require ARCHITECTURE.md diagram updates on drift (#52)

feat: post-run compiles() grader (#51)

feat(mfa/react): add compile-time check to React MFA eval (#53)

fix(compile): auto-run npm install before compile_command (#55)

chore(evals): add compile_command and compiles() grader to CTE evals