docs: add error pages for infrastructure_timeout and agent_process_failed#150
docs: add error pages for infrastructure_timeout and agent_process_failed#150hongyi-chen wants to merge 2 commits into
Conversation
…iled These two platform error codes are defined in warp-server's platformerrors.go but had no corresponding documentation on docs.warp.dev. Add pages following the existing conventions: - agent-process-failed.mdx: agent process exited unexpectedly during task execution (HTTP 500, not retryable, ERROR state) - infrastructure-timeout.mdx: task forcibly terminated after exceeding the maximum allowed runtime (HTTP 500, not retryable, ERROR state) Wire the pages into: - The Errors sidebar group (src/sidebar.ts) - The Platform errors list on the Errors overview index - vercel.json redirects mapping the underscore form to the hyphen form (both with and without trailing slash), matching the pattern used by every other error code Co-Authored-By: Oz <oz-agent@warp.dev>
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
Audit pass following the existing error pages and the docs style guide: - infrastructure_timeout: simplify the lead sentence, make the platform-vs-user-error callout less circular, add a concrete example for what 'task did not report a terminal status' looks like, and tighten the 'How to resolve' steps with specific things to look for in the environment configuration. - agent_process_failed: clarify that this fires after environment setup has completed (so the distinction from environment_setup_failed is obvious from the lead), spell out which environment-setup steps the other error covers, and replace the vague 'check whether the prompt or environment is causing the agent to crash' with concrete triggers (large prompts, memory-heavy tools, degraded environment). Also confirmed coverage is now 1:1 with platformerrors.go: all 18 ErrorCode constants have a matching .mdx page in the errors directory. Co-Authored-By: Oz <oz-agent@warp.dev>
|
I'm starting a first review of this pull request. You can view the conversation on Warp. I completed the review and no human review was requested for this pull request. Comment Powered by Oz |
![oz-for-oss[bot]](https://avatars.githubusercontent.com/in/3450139?s=60&v=4){kind=small}
There was a problem hiding this comment.
Overview
This PR adds documentation for the infrastructure_timeout and agent_process_failed API error codes, wires them into the errors index and sidebar, and adds underscore-to-hyphen redirects in vercel.json. The pages follow the existing error-page structure, the added links point to existing docs targets, and the redirect/sidebar changes match nearby patterns.
Concerns
- No blocking concerns found.
Verdict
Found: 0 critical, 0 important, 0 suggestions
Approve
Comment /oz-review on this pull request to retrigger a review (up to 3 times on the same pull request).
Powered by Oz
1 participant
Summary
Adds documentation pages for two
platformerrors.goerror codes that ship on the public Oz API but had no corresponding page on docs.warp.dev:infrastructure_timeout— task forcibly terminated by the periodic stale-task job after exceeding the maximum allowed runtime.agent_process_failed— agent process exited unexpectedly during task execution (after environment setup completed).Both are platform errors (task state → ERROR), HTTP 500, not retryable, mirroring how they're declared in
warp-server/logic/ai/ambient_agents/platformerrors/platformerrors.go(seeNewInfrastructureTimeoutandNewAgentProcessFailed).Context
This is the docs.warp.dev (Astro Starlight) counterpart to the GitBook PR warpdotdev/gitbook#924. Astro Starlight is now the live docs source, so the same gap had to be filled here. Content follows existing error-page conventions (
authentication-required.mdx,internal-error.mdx,environment-setup-failed.mdx) instead of the GitBook hint syntax /https://docs.warp.dev/...typeURIs.Bot review feedback from the GitBook PR is incorporated:
infrastructure_timeoutto avoid implying the task is blameless ("enforced by Warp's infrastructure rather than by the task's own logic").internal_errorfrom both pages, and toenvironmentsfrominfrastructure_timeout.Changes
src/content/docs/reference/api-and-sdk/troubleshooting/errors/agent-process-failed.mdx(new)src/content/docs/reference/api-and-sdk/troubleshooting/errors/infrastructure-timeout.mdx(new)src/content/docs/reference/api-and-sdk/troubleshooting/errors/index.mdx— list both under Platform errorssrc/sidebar.ts— add to the Errors sidebar group afterinternal-errorvercel.json— add underscore→hyphen 308 redirects for each code (both with and without trailing slash), matching every other error codeValidation
python3 -c "import json; json.load(open('vercel.json'))"passes.platformerrors.go.Conversation: https://staging.warp.dev/conversation/dae12801-fa4e-406c-8d65-c8010ad4cc39
Run: https://oz.staging.warp.dev/runs/019e6b9f-8f5c-7dd1-bd37-e57a8c6b2d1d
This PR was generated with Oz.