儀表板

2026 年不該再有人值班。

讀你的程式碼、看著你的基礎設施，並在你醒來前修好生產環境的代理。

加入候補名單
或

早期存取正陸續開放中。

Polylane 找出你應用程式裡的問題，並自動替你修好。

它學習你執行的一切。 每一個雲端、每一個儲存庫、每一條依賴關係：一張即時的圖。

Topology
Search your cloud resources...
Galaxy
checkout-edge Worker
Critical Critical to your architecture.

Issue hotspot: 4 issues in the last 7 days

Change hotspot: 12 changes in the last 7 days

Cloudflare · coreplane · earth · Compute
hd-prod Hyperdrive
Critical Critical to your architecture.

Change hotspot: 7 changes in the last 7 days

Cloudflare · coreplane · earth · Databases
ingest-events Lambda function
Standard Important to your architecture.

Issue hotspot: 2 issues in the last 7 days

AWS · coreplane-prod · us-east-1 · Compute
payments-db Database
Critical Critical to your architecture.

Issue hotspot: 1 issue in the last 7 days

Change hotspot: 3 changes in the last 7 days

PlanetScale · coreplane · us-east · Databases

它從不停止尋找問題。 問題自己浮現，包括那些沒有人想到要設警示的。

Issues Critical latency degradation in checkout-edge worker Overview

Critical latency degradation in checkout-edge worker

Incident critical Polylane ·Detected 3 hours ago ·Last seen 4 minutes ago ·2 occurrences
OverviewInvestigationMetricsTimelineProperties
checkout-edge › Cloudflare Worker

Critical latency degradation detected in checkout-edge worker: 18x+ P99 latency spikes sustained for 12 minutes

Causal metrics All metrics
Request Duration
414 ms ▲ 43.1σ
Wall Time
416 ms ▲ 43.0σ
CPU Time
14 ms ▲ 2.6σ
Blast radius
Search your cloud resources...
Graph Table
edge-gateway Cloudflare Worker ··· checkout-edge Cloudflare Worker ··· hd-prod Hyperdrive ··· payments-db PlanetScale ··· cart-svc Cloudflare Worker ···
Analysis Copy

Deploy 9f3c2a1 shrank the Hyperdrive pool hd-prod from 50 connections to 5. Under checkout load, requests queue on connection checkout and P99 rises 18× against the 30-minute baseline. Restoring the pool size restores latency.

Tags
service · checkout-edgeprovider · cloudflaredeploy · 9f3c2a1signal · wall_time_p99

它傾聽你既有的每一則警示。 每一則警示都經過分流，真正的事件被調查並修復。

Issues

Issues

Problems detected by Polylane and alerts from your monitoring tools, triaged automatically.

IncidentsNo incidentAll Search issues... Filter
Title Detected by Status
Critical latency degradation in checkout-edge worker Polylane Incident
checkout-edge P99 latency above 2s for 5m Datadog Incident
TypeError: cannot read 'region' of undefined in checkout handler Sentry Triaging
edge-gateway 5xx ratio above 1% for 5m Grafana Cloud No incident
orders-db CPUUtilization above 80% CloudWatch No incident
payments-api webhook retry queue depth above 5,000 Datadog Resolved

它修的是問題，不是症狀。 它找出每一個偵測到的問題的根本原因，並送出 PR 來修復。

github.com/coreplane/payments-api/pull/491

Cap retries on the checkout webhook worker #491

polylane
Open polylane wants to merge 1 commit into main from polylane/autofix/chat/k3x9f2-4e7d21a
Conversation 1 Commits 1 Checks 1 Files changed 2
polylane bot commented 6 minutes ago ···

Retries on the checkout webhook worker were unbounded: a failing delivery re-queued itself forever and amplified load on payments-api. This caps delivery at 5 attempts with exponential backoff and dead-letters the payload after the last one.

What changed

worker/deliver.ts gains MAX_DELIVERY_ATTEMPTS = 5 and backoff between attempts; exhausted payloads land in checkout-webhooks-dlq instead of re-queueing.

Validation

npm test — 214 passed. A forced failing delivery stopped after 5 attempts and appeared in the dead-letter queue.

Root cause · Why it's safe · Out of scope
polylane added commit 4e7d21a Verified
Review required At least 1 approving review is required
ci / test Successful in 3m 12s Details
Review required Waiting on your review: Polylane never merges on its own
Merging is blocked

它回答值班時會被問到的問題。 誰負責這個、改了什麼、是我們的問題嗎：在對話串裡回覆，附上證據。

# incidents
Messages Add canvas Files & links
Polylane Agent 2:14 AM

Critical latency regression on checkout-edge: Wall Time P99 is 18× baseline, sustained since 2:02 AM, and checkout requests are queuing. I'm investigating.

boris 2:15 AM

probably the Hyperdrive change in tonight's deploy?

👀 1
Polylane Agent 2:16 AM

Working on it

  • ✓ Pulled the last 30 minutes of checkout-edge metrics
  • ✓ Correlated with the change records
  • ◐ Testing the Hyperdrive pool hypothesis
  • ○ Draft the fix

status as of 2:17 AM UTC

Polylane Agent 2:19 AM

You're right. Deploy 9f3c2a1 shrank the hd-prod connection pool from 50 to 5, and checkout requests queue behind connections. Confirmed against the pool metrics and the change record. Fix is drafted, coreplane/checkout-edge#142 restores the pool. CI is running; I'll report back here.

Open in Polylane Configure

它把生產環境帶進你的終端機。 你的程式碼代理在更動任何東西之前，先看到正在執行的是什麼。

claude — ~/checkout-edge

> make region a required field on the checkout request schema

●polylane · searchcallers of POST /checkout

●polylane · runToolqueryLogs · request shapes

⏺ Hold on: cart-svc and edge-gateway still send region-less requests, 41,200 in the last 24h. Requiring it now would 400 them both. Safer: default it, migrate the two callers, then require. Are you okay with this plan?

Waiting for your reply…

它記住每一次事件。 下一次永遠不必從零開始。

Settings Memories

Memories

Institutional knowledge from investigations and observations.

All Mine Search memories... Filter 7 memories
Name Source Created
hd-prod must keep 50 connections: at 5, checkout-edge requests queue and P99 rises 18x Investigation 3 hours ago
Deploys that touch checkout-edge's Hyperdrive binding need the pool size checked in review Change analysis 3 hours ago
cart-svc and edge-gateway still send region-less checkout requests; region stays optional until they migrate Investigation yesterday
payments-api webhook retries are capped at 5 with backoff since #491; dead-letters land in checkout-webhooks-dlq Change analysis 2 days ago
payments-db connection spikes follow cart-svc retries, not checkout traffic Observation 6 days ago
checkout-edge P99 sits at 1.4-1.8s on weekday evenings; the Sunday 02:00 UTC bump is the analytics backfill Observation 2 weeks ago
edge-gateway returns 502 for 40s after every deploy while its Fly machines drain; not an incident Change analysis 3 weeks ago

不用遷移。不用新的 SDK。 Polylane 接上你已經在跑的一切。

AWS Cloudflare Vercel Fly.io Render Kubernetes PlanetScale Supabase Modal Railway Convex Datadog Sentry Honeycomb Axiom Grafana Cloud GitHub Slack Cursor Devin Factory MCP 還有更多 →

由值過班的工程師打造。 來自 Cloudflare、Webflow、Groq、Twilio、Uber 和 Robinhood。

Cloudflare Cloudflare Webflow Webflow Groq Groq Twilio Twilio Uber Uber Robinhood Robinhood
「我花了多年打造可觀測性平台，先是在 Baselime， 然後在 Cloudflare。 我們的工具仍然很不擅長找出哪裡壞了，而且沒辦法自己修好任何東西。值班仍然是壞掉的。我正在修好它。」
Boris Tane
Boris Tane
創辦人

我們正在讓軟體自己運作。加入我們。

信任沒有妥協的空間。

AICPA SOC 2

SOC 2 Type II

經獨立稽核的安全控制措施。

ISO 27001

ISO 27001:2022

資訊安全管理的國際標準。

端對端加密

靜態資料採 AES-256，傳輸中採 TLS 1.2+。

資料隔離

每個組織完全隔離。

查看 trust center

常見問題。

Polylane 是什麼？

Polylane 讓你的軟體自主運作。AI 代理學習你執行的一切、監看生產環境的問題、分流你既有的每一則警示、修復能修的部分、回答值班時會被問到的問題、把同樣的生產環境情境交給你的程式碼代理，並記住每一次事件。

Polylane 怎麼偵測問題？

代理讀取真實的指標、日誌與追蹤序列，並對照每個資源平常的行為來判斷。你團隊已經在畫的圖表都會變成檢查；閾值只在它是正確工具的地方存在。

Polylane 會在沒有審查的情況下更動生產環境嗎？

不會，除非你明確允許。復原預設關閉，且限定在你選擇的服務。程式碼變更絕不跳過審查：修復以 pull request 送達，每一次合併都由你的審查與 CI 把關。

Polylane 支援哪些平台？

AWS、Cloudflare、Vercel、Fly.io、Render、Kubernetes、PlanetScale、Supabase 和 Modal，加上 GitHub、Slack 與你的可觀測性工具：Datadog、Honeycomb、Axiom、Grafana Cloud 和 Sentry。

我什麼時候能用 Polylane？

Polylane 尚未正式上市。早期存取正陸續開放中，透過本頁的候補名單申請。

AI 代理能使用 Polylane 嗎？

可以。平台提供位於 api.polylane.com 的 REST API（OpenAPI 規格在 api.polylane.com/v1/doc），以及位於 mcp.polylane.com/mcp 的 MCP 伺服器，支援 OAuth 2.0 動態用戶端註冊；請參閱 polylane.com/auth.md。網站內容的機器可讀版本在 polylane.com/llms.txt。