chore(deps): upgrade eve from 0.11.7 to 0.22.5#88
Merged
Conversation
Vercel now requires a newer minimum eve version for deployment. The only compile-level break across the eleven intervening 0.x minors is the eve 0.15.0 eval assertion overhaul: the run-level success assertion is now t.succeeded() instead of t.completed(). Agent, tool, sandbox, and instrumentation contracts compile unchanged; the eve eval CLI flags embedded in plan artifacts still exist; and the new maxSubagentDepth default of 1 matches the ADR 0015 contract that stage subagents cannot delegate further. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
ncolesummers
added a commit
that referenced
this pull request
Jul 11, 2026
chore(deps): upgrade eve from 0.11.7 to 0.22.5 (re-land #88)
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
Upgrades
evefrom0.11.7to0.22.5(eleven breaking 0.x minors) to satisfy Vercel's new minimum framework version for deployment.The migration surface turned out to be small: the only compile-level break for LoopWorks was the eval assertion API overhaul in eve 0.15.0, which renamed the run-level success assertion. Both evals move from
t.completed()tot.succeeded(); the rest of our assertion vocabulary (calledTool,calledSubagent,notCalledTool,maxToolCalls,noFailedActions,check+eve/evals/expectbuilders) is unchanged in 0.22.Confirmed non-impacts from the 0.12–0.22 changelog review:
defineAgent/defineTool/disableTool/defineSandbox/defaultBackend/defineInstrumentation— our usage typechecks and compiles unchanged;modelOptions.providerOptionsremains a documented, supported path for provider-specific reasoning effort (a top-levelreasoningoption now exists as an alternative).authremoval (0.13.0) and the AI SDK 7approvalstandardization (0.14.0) — LoopWorks tools use neither.limits.maxSubagentDepthdefault dropped from 3 to 1 (0.22.5) — our topology is exactly one delegation level (orchestrator → planner/test-writer), and ADR 0015 requires that stage subagents cannot delegate further, so the new default now enforces our contract.eve evalCLI flags embedded in plan artifacts (--list,--timeout,--skip-report) all still exist.runtimeoption; we usedefaultBackend()withnetworkPolicyonly.Validation
bun run validate(format, lint, agent-docs, markdownlint, typecheck, vitest 52 files / 332 tests, Storybook build, Playwright e2e): PASSbun run build: PASSbunx eve eval --liston 0.22.5 discovers both evals and compiles the full agent tree (root orchestrator, planner and test-writer subagents, tools, sandboxes) with no diagnostics.Notes
Stacked on #87 (
codex/47-test-agent) because the agent tree it introduces is what this upgrade exercises; GitHub will retarget this PR tomainautomatically when #87 merges.🤖 Generated with Claude Code