Skip to content

Backport #2699: fix(ai): recover from invalid tool-call input instead of aborting the agent stream#2703

Merged
VaguelySerious merged 2 commits into
stablefrom
backport/pr-2699-to-stable
Jul 1, 2026
Merged

Backport #2699: fix(ai): recover from invalid tool-call input instead of aborting the agent stream#2703
VaguelySerious merged 2 commits into
stablefrom
backport/pr-2699-to-stable

Conversation

@github-actions

@github-actions github-actions Bot commented Jun 30, 2026

Copy link
Copy Markdown
Contributor

Automated backport of #2699 to stable (backport job run).

AI recommendation: This is a self-contained bug fix to executeTool in packages/ai/src/agent/durable-agent.ts, which exists on stable with the same throw parseError; behavior being fixed (verified at line 1654 on origin/stable). The change makes invalid tool-call input recoverable instead of aborting the stream, adds regression tests, and does not depend on any main-only APIs, so it applies cleanly to stable.

@github-actions @boomyao
fix(ai): recover from invalid tool-call input instead of aborting the…
f4d9c52 
… agent stream (#2699)

In-repo copy of #2192 by @boomyao, opened to run the full CI suite.

Co-authored-by: yao <zhangyaoruo@outlook.com>
Signed-off-by: Peter Wielander <mittgfu@gmail.com>
@github-actions github-actions Bot requested review from a team and ijjk as code owners June 30, 2026 01:45
@github-actions github-actions Bot mentioned this pull request Jun 30, 2026
Merged
@changeset-bot

changeset-bot Bot commented Jun 30, 2026 •
edited
Loading

Copy link
Copy Markdown

🦋 Changeset detected

Latest commit: 25f21a9

The changes in this PR will be included in the next version bump.

This PR includes changesets to release 1 package
Name Type
@workflow/ai Patch

Not sure what this means? Click here to learn what changesets are.

Click here if you're a maintainer who wants to add another changeset to this PR

@vercel

vercel Bot commented Jun 30, 2026 •
edited
Loading

Copy link
Copy Markdown
Contributor

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

Project Deployment Actions Updated (UTC)
example-nextjs-workflow-turbopack Ready Ready Preview, Comment Jul 1, 2026 12:02am
example-nextjs-workflow-webpack Ready Ready Preview, Comment Jul 1, 2026 12:02am
example-workflow Ready Ready Preview, Comment Jul 1, 2026 12:02am
workbench-astro-workflow Ready Ready Preview, Comment Jul 1, 2026 12:02am
workbench-express-workflow Ready Ready Preview, Comment Jul 1, 2026 12:02am
workbench-fastify-workflow Ready Ready Preview, Comment Jul 1, 2026 12:02am
workbench-hono-workflow Ready Ready Preview, Comment Jul 1, 2026 12:02am
workbench-nitro-workflow Ready Ready Preview, Comment Jul 1, 2026 12:02am
workbench-nuxt-workflow Ready Ready Preview, Comment Jul 1, 2026 12:02am
workbench-sveltekit-workflow Ready Ready Preview, Comment Jul 1, 2026 12:02am
workbench-tanstack-start-workflow Ready Ready Preview, Comment Jul 1, 2026 12:02am
workbench-vite-workflow Ready Ready Preview, Comment Jul 1, 2026 12:02am
workflow-docs Ready Ready Preview, Comment, Open in v0 Jul 1, 2026 12:02am
workflow-swc-playground Ready Ready Preview, Comment Jul 1, 2026 12:02am
workflow-tarballs Ready Ready Preview, Comment Jul 1, 2026 12:02am
workflow-web Ready Ready Preview, Comment Jul 1, 2026 12:02am

@github-actions

github-actions Bot commented Jun 30, 2026 •
edited
Loading

Copy link
Copy Markdown
Contributor Author

🧪 E2E Test Results

❌ Some tests failed

Summary

Passed Failed Skipped Total
✅ ▲ Vercel Production 1077 0 78 1155
✅ 💻 Local Development 1174 0 86 1260
✅ 📦 Local Production 1174 0 86 1260
✅ 🐘 Local Postgres 1162 0 98 1260
✅ 🪟 Windows 105 0 0 105
❌ 🌍 Community Worlds 79 102 9 190
✅ 📋 Other 592 0 38 630
Total 5363 102 395 5860

❌ Failed Tests

🌍 Community Worlds (102 failed)

redis (19 failed):

  • hookWorkflow | wrun_01KWFMBB5X7VB9ESJRYWBSPR0K
  • hookWorkflow is not resumable via public webhook endpoint | wrun_01KWFMBHQDGR363ESVXTNZY5NC
  • parallelStepsThenWebhookWorkflow - no hook_conflict from same-tick replay race | wrun_01KWFMBSFZVXXZP3677P8YRYBV
  • sleepingWorkflow | wrun_01KWFMCRNV3XSXV6PRJMMKFFM2
  • outputStreamWorkflow negative startIndex (reads from end)
  • outputStreamWorkflow - getTailIndex and getStreamChunks getTailIndex returns correct index after stream completes
  • outputStreamWorkflow - getTailIndex and getStreamChunks getTailIndex returns -1 before any chunks are written
  • outputStreamWorkflow - getTailIndex and getStreamChunks getStreamChunks returns same content as reading the stream
  • concurrent hook token conflict - two workflows cannot use the same hook token simultaneously | wrun_01KWFMM61E2K1XC88AADJR0BNM
  • hookGetConflictWorkflow - awaiting hook.getConflict() registers hook without payload | wrun_01KWFMMGVZB5XMDYC10WFVJ9QN
  • hookGetConflictThenStepParallelWorkflow - hook.getConflict() continuation step runs alongside other steps | wrun_01KWFMMQFQ3AHWCE67H3BVJA8M
  • hookGetConflictWorkflow - hook.getConflict() resolves with the conflicting run when token is already registered | wrun_01KWFMN5DJPM0JQ6KWK14343SR
  • hookClaimOnlyMutexWorkflow - hook works as a pure run mutex without payload data | wrun_01KWFMNZMNJ6B3Y8N5JD0XTM3K
  • hookAdoptOwnerResultWorkflow - duplicate adopts the owner result via conflict.returnValue | wrun_01KWFMP4D2KAJTS576Q3X4TAW5
  • hookSignalOwnerWorkflow - duplicate forwards its payload to the owner via resumeHook | wrun_01KWFMP9MJWPSFWZ5SK9TW8BQX
  • hookSupersedeOwnerWorkflow - duplicate cancels the owner and claims the released token | wrun_01KWFMPF8CMWSQPBSNBVHD0HFR
  • resume-or-start route pattern - resumeHook retried after start() reaches the new run | wrun_01KWFMPQQMJWT7VF186THZ47RX
  • pages router sleepingWorkflow via pages router
  • resilient start: addTenWorkflow completes when run_created returns 500 | wrun_01KWFMW1C9JEPVJKYQES2XF5EA

turso (83 failed):

  • addTenWorkflow | wrun_01KWFMAD5XYBYEC40E7ACBFKWG
  • addTenWorkflow | wrun_01KWFMAD5XYBYEC40E7ACBFKWG
  • deploymentId: 'latest' is a no-op in non-Vercel worlds
  • wellKnownAgentWorkflow (.well-known/agent) | wrun_01KWFMAHSPX2766XH0S4D0BB6R
  • should work with react rendering in step
  • promiseAllWorkflow | wrun_01KWFMAMVRCAH0R409VRT66GN0
  • promiseRaceWorkflow | wrun_01KWFMATSYCJAC60NQCKNNBSMX
  • promiseAnyWorkflow | wrun_01KWFMAXE85CMTE8T42H4F2E26
  • importedStepOnlyWorkflow | wrun_01KWFMAZ0Y453YQZN8WVGWHDGY
  • readableStreamWorkflow | wrun_01KWFMAZF2Q84CDS51J0XCZDCG
  • hookWorkflow | wrun_01KWFMBB5X7VB9ESJRYWBSPR0K
  • hookWorkflow is not resumable via public webhook endpoint | wrun_01KWFMBHQDGR363ESVXTNZY5NC
  • webhookWorkflow | wrun_01KWFMBNFFJ3E8QR77DXRJ968D
  • parallelStepsThenWebhookWorkflow - no hook_conflict from same-tick replay race | wrun_01KWFMBSFZVXXZP3677P8YRYBV
  • sleepingWorkflow | wrun_01KWFMCRNV3XSXV6PRJMMKFFM2
  • parallelSleepWorkflow | wrun_01KWFMD82S5TF0Q4728Z8RN39S
  • sleepWinsRaceWorkflow | wrun_01KWFMDB6G3EPHKX2GDC9YCGG2
  • stepWinsRaceWorkflow | wrun_01KWFMDED2B9Q6GKNRP3PYCG79
  • nullByteWorkflow | wrun_01KWFMDHJ3GKFD3XHZSKNWFGRK
  • workflowAndStepMetadataWorkflow | wrun_01KWFMDKK00RF62GRKM6XF7SV0
  • outputStreamWorkflow no startIndex (reads all chunks)
  • outputStreamWorkflow positive startIndex (skips first chunk)
  • outputStreamWorkflow negative startIndex (reads from end)
  • outputStreamWorkflow - getTailIndex and getStreamChunks getTailIndex returns correct index after stream completes
  • outputStreamWorkflow - getTailIndex and getStreamChunks getTailIndex returns -1 before any chunks are written
  • outputStreamWorkflow - getTailIndex and getStreamChunks getStreamChunks returns same content as reading the stream
  • outputStreamInsideStepWorkflow - getWritable() called inside step functions | wrun_01KWFMFW8ZQD1HSN34VFFG0ZCJ
  • writableForwardedFromWorkflowWorkflow | wrun_01KWFMG9T5FMSSSDMXMX2RKC89
  • writableForwardedFromStepWorkflow | wrun_01KWFMGE7Q7H3C47MRWQMYC4B8
  • fetchWorkflow | wrun_01KWFMGHHSRJWFHP2MJJEHX1EC
  • promiseRaceStressTestWorkflow | wrun_01KWFMGMKSSWVMAYT55FW2YB8P
  • error handling error propagation workflow errors nested function calls preserve message and stack trace
  • error handling error propagation workflow errors cross-file imports preserve message and stack trace
  • error handling error propagation step errors basic step error preserves message and stack trace
  • error handling error propagation step errors cross-file step error preserves message and function names in stack
  • error handling retry behavior regular Error retries until success
  • error handling retry behavior FatalError fails immediately without retries
  • error handling retry behavior RetryableError respects custom retryAfter delay
  • error handling retry behavior maxRetries=0 disables retries
  • error handling catchability FatalError can be caught and detected with FatalError.is()
  • error handling not registered WorkflowNotRegisteredError fails the run when workflow does not exist
  • error handling not registered StepNotRegisteredError fails the step but workflow can catch it
  • error handling not registered StepNotRegisteredError fails the run when not caught in workflow
  • hookCleanupTestWorkflow - hook token reuse after workflow completion | wrun_01KWFMKTD3VZ6AXDXE3E3NN04S
  • concurrent hook token conflict - two workflows cannot use the same hook token simultaneously | wrun_01KWFMM61E2K1XC88AADJR0BNM
  • hookGetConflictWorkflow - awaiting hook.getConflict() registers hook without payload | wrun_01KWFMMGVZB5XMDYC10WFVJ9QN
  • 'hookGetConflictWithPriorStepWorkflow' - hook.getConflict() does not block step execution | wrun_01KWFMMK6MMTWMJ4T7RFZHAYVA
  • 'hookGetConflictWithParallelStepWorkfl…' - hook.getConflict() does not block step execution | wrun_01KWFMMNBN5HRS4RE3BQNWSBJR
  • hookGetConflictThenStepParallelWorkflow - hook.getConflict() continuation step runs alongside other steps | wrun_01KWFMMQFQ3AHWCE67H3BVJA8M
  • hookGetConflictWorkflow - hook.getConflict() resolves with the conflicting run when token is already registered | wrun_01KWFMN5DJPM0JQ6KWK14343SR
  • hookClaimOnlyMutexWorkflow - hook works as a pure run mutex without payload data | wrun_01KWFMNZMNJ6B3Y8N5JD0XTM3K
  • hookAdoptOwnerResultWorkflow - duplicate adopts the owner result via conflict.returnValue | wrun_01KWFMP4D2KAJTS576Q3X4TAW5
  • hookSignalOwnerWorkflow - duplicate forwards its payload to the owner via resumeHook | wrun_01KWFMP9MJWPSFWZ5SK9TW8BQX
  • hookSupersedeOwnerWorkflow - duplicate cancels the owner and claims the released token | wrun_01KWFMPF8CMWSQPBSNBVHD0HFR
  • resume-or-start route pattern - resumeHook retried after start() reaches the new run | wrun_01KWFMPQQMJWT7VF186THZ47RX
  • hookDisposeTestWorkflow - hook token reuse after explicit disposal while workflow still running | wrun_01KWFMPZ3KYJYV5Z77JYMSXHBN
  • stepFunctionPassingWorkflow - step function references can be passed as arguments (without closure vars) | wrun_01KWFMQCZ17RHK1SNS5Q0MJMNK
  • stepFunctionWithClosureWorkflow - step function with closure variables passed as argument | wrun_01KWFMQN4FAWD06BBZ9A9588SV
  • closureVariableWorkflow - nested step functions with closure variables | wrun_01KWFMQT9VWB15GE4H9HDTDBV4
  • spawnWorkflowFromStepWorkflow - spawning a child workflow using start() inside a step | wrun_01KWFMQW7NM3DP4S6WYFPV3DEM
  • health check (queue-based) - workflow and step endpoints respond to health check messages
  • health check (CLI) - workflow health command reports healthy endpoints
  • pathsAliasWorkflow - TypeScript path aliases resolve correctly | wrun_01KWFMRAXDVH07EPJZ1V2R6YRD
  • Calculator.calculate - static workflow method using static step methods from another class | wrun_01KWFMRG2ZAJR7AP4MV2891H02
  • AllInOneService.processNumber - static workflow method using sibling static step methods | wrun_01KWFMRPJBFNMGYJBAKD8R7YV1
  • ChainableService.processWithThis - static step methods using this to reference the class | wrun_01KWFMRWVRKSGKN8NT2G5T5769
  • thisSerializationWorkflow - step function invoked with .call() and .apply() | wrun_01KWFMS39SXVDQTXE2194SAXTQ
  • customSerializationWorkflow - custom class serialization with WORKFLOW_SERIALIZE/WORKFLOW_DESERIALIZE | wrun_01KWFMSAPH7TJM54FPP2GFBARX
  • instanceMethodStepWorkflow - instance methods with "use step" directive | wrun_01KWFMSGTM40HW0Y7R8XXMY329
  • crossContextSerdeWorkflow - classes defined in step code are deserializable in workflow context | wrun_01KWFMSXAYDGQ5GDXWGMWDFB24
  • stepFunctionAsStartArgWorkflow - step function reference passed as start() argument | wrun_01KWFMT5Z80EFB6HDH3K3ZDSZC
  • cancelRun - cancelling a running workflow | wrun_01KWFMTC877KDSQZ8GPJJHW0S3
  • cancelRun via CLI - cancelling a running workflow | wrun_01KWFMTGHS0BSXD3V2S64QA77N
  • pages router addTenWorkflow via pages router
  • pages router promiseAllWorkflow via pages router
  • pages router sleepingWorkflow via pages router
  • hookWithSleepWorkflow - hook payloads delivered correctly with concurrent sleep | wrun_01KWFMTQEVWE2QK20DXMGP88DT
  • hookWithSleepFinalStepWorkflow - step only on final payload | wrun_01KWFMV2H00SHFHVAVQJ191GQ0
  • sleepInLoopWorkflow - sleep inside loop with steps actually delays each iteration | wrun_01KWFMVAK7CEGR0QRPFHK2GGXR
  • sleepWithSequentialStepsWorkflow - sequential steps work with concurrent sleep (control) | wrun_01KWFMVNMXBRKG5X5Q82P17CMY
  • importMetaUrlWorkflow - import.meta.url is available in step bundles | wrun_01KWFMVX27MJ4ER45NS59WWJMA
  • metadataFromHelperWorkflow - getWorkflowMetadata/getStepMetadata work from module-level helper (#1577) | wrun_01KWFMVZ9VMXZYEQ26PHXTN9AD
  • resilient start: addTenWorkflow completes when run_created returns 500 | wrun_01KWFMW1C9JEPVJKYQES2XF5EA

Details by Category

✅ ▲ Vercel Production
App Passed Failed Skipped
✅ astro 97 0 8
✅ example 97 0 8
✅ express 97 0 8
✅ fastify 97 0 8
✅ hono 97 0 8
✅ nextjs-turbopack 102 0 3
✅ nextjs-webpack 102 0 3
✅ nitro 97 0 8
✅ nuxt 97 0 8
✅ sveltekit 97 0 8
✅ vite 97 0 8
✅ 💻 Local Development
App Passed Failed Skipped
✅ astro-stable 99 0 6
✅ express-stable 99 0 6
✅ fastify-stable 99 0 6
✅ hono-stable 99 0 6
✅ nextjs-turbopack-canary 86 0 19
✅ nextjs-turbopack-stable 105 0 0
✅ nextjs-webpack-canary 86 0 19
✅ nextjs-webpack-stable 105 0 0
✅ nitro-stable 99 0 6
✅ nuxt-stable 99 0 6
✅ sveltekit-stable 99 0 6
✅ vite-stable 99 0 6
✅ 📦 Local Production
App Passed Failed Skipped
✅ astro-stable 99 0 6
✅ express-stable 99 0 6
✅ fastify-stable 99 0 6
✅ hono-stable 99 0 6
✅ nextjs-turbopack-canary 86 0 19
✅ nextjs-turbopack-stable 105 0 0
✅ nextjs-webpack-canary 86 0 19
✅ nextjs-webpack-stable 105 0 0
✅ nitro-stable 99 0 6
✅ nuxt-stable 99 0 6
✅ sveltekit-stable 99 0 6
✅ vite-stable 99 0 6
✅ 🐘 Local Postgres
App Passed Failed Skipped
✅ astro-stable 98 0 7
✅ express-stable 98 0 7
✅ fastify-stable 98 0 7
✅ hono-stable 98 0 7
✅ nextjs-turbopack-canary 85 0 20
✅ nextjs-turbopack-stable 104 0 1
✅ nextjs-webpack-canary 85 0 20
✅ nextjs-webpack-stable 104 0 1
✅ nitro-stable 98 0 7
✅ nuxt-stable 98 0 7
✅ sveltekit-stable 98 0 7
✅ vite-stable 98 0 7
✅ 🪟 Windows
App Passed Failed Skipped
✅ nextjs-turbopack 105 0 0
❌ 🌍 Community Worlds
App Passed Failed Skipped
✅ mongodb-dev 3 0 3
✅ redis-dev 3 0 3
❌ redis 67 19 0
✅ turso-dev 3 0 3
❌ turso 3 83 0
✅ 📋 Other
App Passed Failed Skipped
✅ e2e-local-dev-nest-stable 99 0 6
✅ e2e-local-dev-tanstack-start-stable 99 0 6
✅ e2e-local-postgres-nest-stable 98 0 7
✅ e2e-local-postgres-tanstack-start-stable 98 0 7
✅ e2e-local-prod-nest-stable 99 0 6
✅ e2e-local-prod-tanstack-start-stable 99 0 6

📋 View full workflow run

@NathanColosimo NathanColosimo mentioned this pull request Jun 30, 2026
Closed
@VaguelySerious VaguelySerious enabled auto-merge (squash) July 1, 2026 00:04
@VaguelySerious VaguelySerious merged commit ef60d31 into stable Jul 1, 2026
144 of 162 checks passed
@VaguelySerious VaguelySerious deleted the backport/pr-2699-to-stable branch July 1, 2026 20:12
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@VaguelySerious VaguelySerious VaguelySerious approved these changes
@ijjk ijjk Awaiting requested review from ijjk ijjk is a code owner

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

1 participant

@VaguelySerious