Skip to content

fix(eve): fill connection call defaults once, when the call runs - #5511

Merged
AndrewBarba merged 2 commits into
triage/5509-null-default-reprofrom
triage/5509-null-default-fix
Oct 29, 2026
Merged

AndrewBarba merged 2 commits into
triage/5509-null-default-reprofrom
triage/5509-null-default-fix

Conversation

@roboeve

@roboeve roboeve Bot commented Oct 28, 2026 •

Copy link
Copy Markdown
Contributor

Fixes #5509: connection tool calls no longer reject a schema default that eve filled in itself, such as the null Pydantic publishes for x: str = None.

Why

Since #4400 (eve@0.76.0), a connection call is checked twice: once by the entry's input schema before approval (checkInput), and again when the call runs (checkCall in callConnectionTool). The first check returned the input with schema defaults filled in, and the harness ran the tool on that value. applyDefaults fills a default even when its own schema rejects it (on purpose, see #3696), so for { "type": "string", "default": null } the second check rejected the null the first check added. The call failed with query: Instance type "null" is invalid. Expected "string". and never reached the server. Azure Managed Grafana's MCP server and other Pydantic-based servers publish many of these.

Approach

Because the pre-approval check returned a transformed value that the call validates again, returning the model's input unchanged from that check (minus provided arguments, as before) should leave the call's check as the only place defaults are filled. That restores what eve@0.75.1 did, which validated connection input once, when the call ran.

  • checkInput still rejects a bad call before anyone is asked to approve it, but it now returns the model's input instead of the default-filled one.
  • Dropping toolCall.providedArguments keys moves into withoutProvidedArguments, which both checks call, so the approval prompt still never shows a model value for a provided key (fix(eve): drop model values for provided connection arguments before validation #4652).
  • checkCall now only looks up the tool, validates, and fills defaults.

The fillProperties change the issue suggests would also fix this case, but it changes defineJsonSchema for every JSON Schema tool and drops a documented behavior. It would also leave the double check in place for any other default that does not pass a second validation.

Behavior change: the call input that reaches the approval policy and the session history no longer includes schema defaults. Before #4400 it did not include them either.

Validation

  • The repro test from fix(eve): fill connection call defaults once, when the call runs #5510 now passes without it.fails. It asserts that the server receives { query: null } and that the pre-approval check returns {}.
  • step-catalog.test.ts: the eve__call validation case now expects { team: "core" } from the pre-approval check, while the call still sends { limit: 20, team: "core" } to the server. The fix(eve): drop model values for provided connection arguments before validation #4652 provided-argument case is unchanged and passes.
  • kennel-mcp-results eval: the kennel__list_feedings check no longer matches day: "today" in the recorded call input, which now holds the model's { petId: 4217 }. The fixture server echoes the day it receives, and the eval's output check still requires today there, so it still proves the default reaches the server. Against a local mock-model build, the list_feedings gate passes. The only local failure was photo-image-parts, because this environment has no Docker sandbox provider. That gate passed in CI on the previous head.
  • pnpm --filter eve exec vitest run --config vitest.unit.config.ts src/execution/catalog/: 125 passed.
  • Connection integration tests (tool-stubs/catalog, harness/tools-schema, connections/events): 56 passed.
  • tsc -p packages/eve/tsconfig.json --noEmit, oxfmt --check, and oxlint pass.

A connection call was checked before approval and again at execution, and the second check validated the first one's default-filled output. A default its own schema rejects, such as Pydantic's null for x: str = None, failed that second check. The pre-approval check now returns the model's input without provided arguments, and only the execution check fills defaults.

Fixes #5509

Co-authored-by: ruiconti <1834568+ruiconti@users.noreply.github.com>
@vercel

vercel Bot commented Oct 28, 2026 •

Copy link
Copy Markdown
Contributor

The latest updates on your projects. Learn more about Vercel for GitHub.

Project Deployment Actions Updated
eve-docs Ready Ready Preview, v0 Oct 28, 2026 1:10pm UTC
1 Skipped Deployment
Project Deployment Actions Updated
eve-pkg Skipped Skipped v0 Oct 28, 2026 1:10pm UTC

@vercel-internal-security-reviewer

Copy link
Copy Markdown

Security review details

@github-actions

github-actions Bot commented Oct 28, 2026 •

Copy link
Copy Markdown
Contributor

Bundle + Package Summary: apps/fixtures/weather-agent

Key takeaways

  • No notable deltas vs triage/5509-null-default-repro (58ebcac).

Delta vs triage/5509-null-default-repro (58ebcac)

Area Metric Baseline Current Delta
Package Packed tarball 7.93 MB 7.93 MB -130 B ✅
Package Unpacked publish size 27.69 MB 27.69 MB +103 B ⚠️
Package Installed footprint 76.09 MB 76.09 MB +103 B ⚠️
Package Published files 4385 4385 0
Package Installed files 8382 8382 0
Package Installed package instances 34 34 0
Package Distinct installed package names 32 32 0
Package Installed dependency edges 51 51 0
Package Installed optional peer edges 17 17 0
Runtime Unique function payloads 2 2 0
Runtime Total function bytes 22.19 MB 22.19 MB +148 B ⚠️
Runtime Public routes 18 18 0
Changed function payloads vs triage/5509-null-default-repro (58ebcac) (2)
Function Status Baseline Current Delta Route changes
functions/__server.func changed 11.09 MB 11.09 MB +74 B ⚠️ none
functions/.well-known/workflow/v1/flow.func changed 11.10 MB 11.10 MB +74 B ⚠️ none

eve init install

Metric Baseline Current Delta
Installed footprint 114.47 MB 114.47 MB +103 B ⚠️
Installed packages 97 97 0
dependencies 4 4 0
devDependencies 2 2 0
Dependency package bytes 46.17 MB 46.17 MB +103 B ⚠️
devDependency package bytes 5.11 MB 5.11 MB 0 B ➖
Build Metadata
  • Preset: vercel
  • Nitro: nitro@3.0.260903-beta
  • Output directory: apps/fixtures/weather-agent/.vercel/output
  • Build metadata timestamp: 2026-10-28T13:10:35.387Z
  • Route aliases: 18 public, 1 internal (19 total aliases)
  • Vercel routes in config: 21
  • Severity legend: 🔴 dominant/large, 🟠 notable, 🟡 watch, ⚪ small
Package Drill-Down

Package Details

  • Package: eve@0.85.0
  • Package directory: packages/eve
  • Tarball: 7.93 MB (eve-0.85.0.tgz)
  • Unpacked payload: 27.69 MB across 4385 published files
  • Installed footprint: 76.09 MB across 8382 installed files
  • Installed root package: 27.69 MB
  • Installed dependencies: 48.40 MB
  • Installed package instances: 34
  • Distinct installed package names: 32
  • Installed dependency edges: 51
  • Installed optional peer edges: 17
  • Runtime dependencies: 2
  • Peer dependencies: 15 (14 optional)

Installed footprint is measured from an isolated temporary npm install of the packed tarball.
Graph metrics read only package.json files in package directories directly beneath a node_modules boundary, including nested boundaries. Each directory is one package instance; distinct names come from those manifests. Dependency edges count each unique name in dependencies or optionalDependencies per instance; optional peer edges count peerDependencies marked optional.

Heavy installed dependencies

  • eve: 27.69 MB (36.4%)
  • @rolldown/binding-linux-x64-gnu: 19.89 MB (26.1%)
  • ai: 8.45 MB (11.1%)
  • zod: 6.15 MB (8.1%)
  • undici: 3.58 MB (4.7%)
Publish payload breakdown
Published file size
🟠 dist/src/compiled/shadcn-registry/index.js       [#####...................] 3.85 MB 13.9%
🟠 dist/src/compiled/@photon-ai/chat-adapter-ime... [###.....................] 2.29 MB 8.3%
🟠 dist/src/compiled/run/index.js                   [#.......................] 1.00 MB 3.6%
🟡 dist/src/compiled/@vercel/blob/index.js          [#.......................] 604.9 kB 2.2%
🟡 dist/src/compiled/_chunks/workflow/undici-D-I... [#.......................] 521.9 kB 1.9%
🔴 Other published files                            [########################] 19.42 MB 70.1%
Installed footprint breakdown
Installed package size
🔴 eve                             [########################] 27.69 MB 36.4%
🔴 @rolldown/binding-linux-x64-gnu [#################.......] 19.89 MB 26.1%
🔴 ai                              [#######.................] 8.45 MB 11.1%
🔴 zod                             [#####...................] 6.15 MB 8.1%
🟠 undici                          [###.....................] 3.58 MB 4.7%
🟠 nitro                           [##......................] 1.89 MB 2.5%
🔴 Other installed packages        [#######.................] 8.45 MB 11.1%
Runtime dependencies (2)
Package Range Notes
nitro 3.0.260903-beta
undici 8.10.2
Peer dependencies (15)
Package Range Notes
@linqapp/sdk ^0.47.1 optional peer
@nuxt/kit ^4.0.0 optional peer
@nuxt/schema ^4.0.0 optional peer
@opentelemetry/api ^1.0.0 optional peer
@vercel/connect ^2.5.2 optional peer
ai catalog:
braintrust ^3.0.0 optional peer
chat ^4.42.0 optional peer
dd-trace ^6.13.0 optional peer
just-bash ^3.6.0 optional peer
next ^16.0.0 optional peer
react ^19.0.0 optional peer
svelte ^5.0.0 optional peer
vite ^8.3.1 optional peer
vue ^3.5.0 optional peer
eve init install drill-down

eve init install details

  • Command: eve init my-agent
  • Package manager: npm
  • Installed footprint: 114.47 MB across 10299 installed files
  • Installed packages: 97 total (91 transitive-only)
  • dependencies: 4 direct packages totaling 46.17 MB
  • devDependencies: 2 direct packages totaling 5.11 MB
  • Other transitive package files: 63.20 MB

Installed footprint is measured from an isolated temporary eve init my-agent using the current packed eve tarball.

Heavy installed dependencies

  • @typescript/typescript-linux-x64: 27.95 MB (24.4%)
  • eve: 27.69 MB (24.2%)
  • @rolldown/binding-linux-x64-gnu: 19.89 MB (17.4%)
  • zod: 9.76 MB (8.5%)
  • ai: 8.45 MB (7.4%)
Installed footprint breakdown
Installed package size
🔴 @typescript/typescript-linux-x64 [########################] 27.95 MB 24.4%
🔴 eve                              [########################] 27.69 MB 24.2%
🔴 @rolldown/binding-linux-x64-gnu  [#################.......] 19.89 MB 17.4%
🔴 zod                              [########................] 9.76 MB 8.5%
🔴 ai                               [#######.................] 8.45 MB 7.4%
🟠 undici                           [###.....................] 3.58 MB 3.1%
🔴 Other installed packages         [###############.........] 17.17 MB 15.0%
dependencies (4)
Package Range Installed size Share
@vercel/connect 2.5.2 276.3 kB 0.2%
ai ^7.0.137 8.45 MB 7.4%
eve file:eve-0.85.0.tgz 27.69 MB 24.2%
zod 4.5.4 9.76 MB 8.5%
devDependencies (2)
Package Range Installed size Share
@types/node 24.x 2.61 MB 2.3%
typescript 7.0.2 2.50 MB 2.2%
Function Drill-Down

Payload Size Graph

Unique function payload size and share of total
🔴 functions/.well-known/workflow/v1/flow.func     [########################] 11.10 MB 50.0%
🔴 functions/__server.func                         [########################] 11.09 MB 50.0%

Top Function Payloads

🟠 functions/.well-known/workflow/v1/flow.func • 1 public route • 11.10 MB
Metric Value
Public routes /.well-known/workflow/v1/flow
Runtime nodejs24.x
Handler index.mjs
Payload 11.10 MB
Function files 11.10 MB across 124 files
Traced dependencies 0 B
Signal 🟠 Bundled file _chunks/vercel.web.mjs is 2.76 MB (24.9%)

🟠 🔎 Dependency Analysis

📦 Bundled files:

Bundled file size
🟠 _chunks/vercel.web.mjs                         [#############...........] 2.76 MB 24.9%
🟡 _libs/undici.mjs                               [#####...................] 994.5 kB 9.0%
🟡 _chunks/sandbox2.mjs                           [#####...................] 937.1 kB 8.4%
🟡 _chunks/compiled-artifacts-instrumentation.mjs [####....................] 730.7 kB 6.6%
🟡 _chunks/token-util-Bla5Z0G9.mjs                [###.....................] 704.5 kB 6.3%
🔴 Other bundled files                            [########################] 4.97 MB 44.8%

🧾 Vercel Config

{
  "handler": "index.mjs",
  "launcherType": "Nodejs",
  "shouldAddHelpers": false,
  "supportsResponseStreaming": true,
  "runtime": "nodejs24.x",
  "maxDuration": "max",
  "experimentalTriggers": [
    {
      "type": "queue/v2beta",
      "topic": "__eve776561746865722d6167656e74_wkf_workflow_*",
      "consumer": "default",
      "retryAfterSeconds": 5,
      "initialDelaySeconds": 0
    }
  ],
  "environment": {
    "WORKFLOW_PRECONDITION_GUARD": "1"
  }
}

🟠 functions/__server.func • 17 public routes, 1 internal alias • 11.09 MB
Metric Value
Public routes /
/.well-known/workflow/v1/webhook/[token]
/eve/v1/callback/[token]
/eve/v1/connections/[name]/callback/[attemptId]/[token]
/eve/v1/connections/[name]/callback/[token]
/eve/v1/health
/eve/v1/info
/eve/v1/session
/eve/v1/session/[parentSessionId]/subagents/[callId]/[childSessionId]/stream
/eve/v1/session/[sessionId]
/eve/v1/session/[sessionId]/cancel
/eve/v1/session/[sessionId]/clear
/eve/v1/session/[sessionId]/compact
/eve/v1/session/[sessionId]/descendants/stream
/eve/v1/session/[sessionId]/reset
/eve/v1/session/[sessionId]/stream
/eve/v1/session/[sessionId]/stubs
Internal aliases /__server
Runtime nodejs24.x
Handler index.mjs
Payload 11.09 MB
Function files 11.09 MB across 124 files
Traced dependencies 0 B
Signal 🟠 Bundled file _chunks/vercel.web.mjs is 2.76 MB (24.9%)

🟠 🔎 Dependency Analysis

📦 Bundled files:

Bundled file size
🟠 _chunks/vercel.web.mjs                         [#############...........] 2.76 MB 24.9%
🟡 _libs/undici.mjs                               [#####...................] 994.5 kB 9.0%
🟡 _chunks/sandbox2.mjs                           [#####...................] 937.1 kB 8.4%
🟡 _chunks/compiled-artifacts-instrumentation.mjs [####....................] 730.7 kB 6.6%
🟡 _chunks/token-util-Bla5Z0G9.mjs                [###.....................] 704.5 kB 6.3%
🔴 Other bundled files                            [########################] 4.97 MB 44.8%

🧾 Vercel Config

{
  "handler": "index.mjs",
  "launcherType": "Nodejs",
  "shouldAddHelpers": false,
  "supportsResponseStreaming": true,
  "runtime": "nodejs24.x"
}

Build Timing: e2e/fixtures/agent-tools-sandbox

This is an informational timing measurement inside eve build, from preflight through publication. Output-size measurement and profile writing are excluded.

Build mode: deployable Vercel build with sandbox template prewarm included.

  • Build pipeline: 3.76 s -> 8.97 s (+5.21 s) vs triage/5509-null-default-repro (58ebcac).
  • Timing is informational: shared GitHub runners are too variable for a hard timing budget.
Detailed phase timings vs `triage/5509-null-default-repro (58ebcac)`
Phase Baseline Current Delta
extension.check 20.5 ms 9.3 ms -11.2 ms
project.resolve 6.1 ms 0.4 ms -5.7 ms
workspace.create 1.1 ms 0.6 ms -0.5 ms
host.prepare 572.1 ms 692.8 ms +120.7 ms
vercel.service-prefix.resolve 1.8 ms 1.8 ms 0.0 ms
nitro.create 463.3 ms 549.1 ms +85.8 ms
sandbox.prewarm 457.6 ms 5.33 s +4.88 s
nitro.cache.prepare 0.3 ms 0.3 ms 0.0 ms
nitro.prepare 0.8 ms 0.9 ms +0.1 ms
nitro.public-assets 0.8 ms 0.8 ms 0.0 ms
nitro.prerender 0.5 ms 0.5 ms 0.0 ms
nitro.bundle 2.08 s 2.18 s +99.4 ms
nitro.cache.write 0.4 ms 0.6 ms +0.2 ms
vercel.workflow-function.materialize 64.6 ms 63.1 ms -1.5 ms
agent-summary.emit 0.7 ms 0.7 ms 0.0 ms
connect-manifest.emit 0.3 ms 0.3 ms 0.0 ms
nitro.close 0.1 ms 0.1 ms 0.0 ms
output.publish 3.7 ms 3.6 ms -0.1 ms
workspace.remove 2.8 ms 2.8 ms 0.0 ms

The call input now holds the model's arguments, so the eval reads the schema default from the day the fixture server echoes.

Co-authored-by: ruiconti <1834568+ruiconti@users.noreply.github.com>
@vercel
vercel Bot temporarily deployed to Preview – eve-pkg October 28, 2026 13:07 Inactive
@roboeve
roboeve Bot requested a review from ruiconti October 28, 2026 13:07
@roboeve

roboeve Bot commented Oct 28, 2026

Copy link
Copy Markdown
Contributor Author

CI on 8974bcd failed in three places:

  • agent-openapi-swagger e2e (Postgres and Vercel): the kennel-mcp-results eval expected the recorded kennel__list_feedings input to include the filled day: "today" default, but the call now records only {"petId":4217}. 00444ef already fixes this by checking the default through the server echo. I reproduced the old failure locally and the eval passes 8/8 with an equivalent change.
  • test-webchat (usage-limit prompt timed out after a 500) and test-integration (windows-latest) (session-stranded.integration.test.ts:197) don't touch connection code and look like flakes. The new head reruns them.

Correction to the PR body: "Before #4400 it did not include them either" is only half true. In eve@0.75.1 the approval input had no defaults, but the reported nested call input did include them. That's why this eval asserted input: { day: "today" } before #4400. So the session-history part of this PR is a real behavior change compared with 0.75.1, not a restoration of it.

@roboeve
roboeve Bot marked this pull request as ready for review October 28, 2026 13:26
@roboeve
roboeve Bot requested a review from a team as a code owner October 28, 2026 13:26
@AndrewBarba
AndrewBarba removed this pull request from stack #5512 October 29, 2026 17:46
@AndrewBarba
AndrewBarba merged commit d72750d into triage/5509-null-default-repro Oct 29, 2026
195 of 197 checks passed
@AndrewBarba
AndrewBarba deleted the triage/5509-null-default-fix branch October 29, 2026 17:46
@AndrewBarba

Copy link
Copy Markdown
Collaborator

Thanks, agreed. In eve@0.75.1 the recorded call input included defaults, so the session-history part is a behavior change, not a restoration. The combined description on #5510 now states it that way.

This branch was successfully deployed

1 active and 1 inactive deployments
Preview – eve-docs — 00444ef3 Deployed Oct 28, 2026 by vercel[bot]
Preview – eve-pkg — 00444ef3 Deployed Oct 28, 2026 by vercel[bot]
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Connection tool calls reject their own default: null since 0.76.0

2 participants