{"id":"devopsbench","measured_tasks":0,"name":"DevOpsBench-100","note":"Complete source roster. Preparation is not execution, grading or a leaderboard score.","official_tasks":100,"schema_version":1,"scope":"full","source_sha256":"1efd654cd3248d225964702c8b4c6f7c9f3909a557b7a9c64eee91642e12efce","source_url":"https://blobfish.ai/benchmarks/devopsbench-100","split":"all 100 released tasks; original train/heldout labels preserved","status":"prepared","task_manifest_sha256":"a6ba2fd08d75c8c5ce7945c56a53242aba2df361ffedce5b73bb7764d2fb4a3d","tasks":[{"category":"aiops_analysis","id":"dob100-001-rca-analytics-egress-blocked","prompt":"Analytics connections are refused, not slow. The incident record describes the visible effect, but it does not establish whether the apparent owner, a dependency, or the deployed configuration is responsible. The decision affects reporting freshness, warehouse replica access, scheduled rollups, and downstream decision data. The disputed mechanism is outbound connectivity, DNS reachability, firewall policy, and dependency access. Explain which conclusion survives cross-checking, record it in the existing incident work item, and leave the live evidence that lets the incoming engineer reproduce it; production controls should remain unchanged. This diagnosis will decide whether analytics worker enters recovery on 2026-03-04. Before closing the incident record, give the incoming owner the earliest credible recovery date, the constraint behind it, and whether approved acceleration or an exception would materially help.\n","reference_tool_calls":34,"source_split":"train","source_task_digest":"f13c791d55aa579013fd4e1684357eedb6684ba08c158b00f2c64006fc255bf0"},{"category":"aiops_analysis","id":"dob100-002-rca-analytics-revoked-grant","prompt":"Analytics cannot read the replica since the rotation. The incident record describes the visible effect, but it does not establish whether the apparent owner, a dependency, or the deployed configuration is responsible. The handoff has to remain sound across reporting freshness, warehouse replica access, scheduled rollups, and downstream decision data. The disputed mechanism is database authorization, revoked roles, credential scope, and least-privilege access. Resolve the disagreement far enough for the incident lead to act, keep production unchanged, and record both the supported conclusion and any uncertainty the evidence cannot settle. Analytics worker is tentatively booked for recovery on 2026-03-04; the evidence here determines whether that promise is credible. Say whether the date holds against current on-call commitments and change windows, what a sanctioned acceleration costs, and which faster route still lacks authority.\n","reference_tool_calls":34,"source_split":"train","source_task_digest":"172eba4cf48a7977f1d32de87ab0f16f359d5f9adba404f46792a70515784485"},{"category":"aiops_analysis","id":"dob100-003-rca-checkout-unschedulable-replicas","prompt":"Checkout is running at a fraction of capacity. The incident record describes the visible effect, but it does not establish whether the apparent owner, a dependency, or the deployed configuration is responsible. The service owner is accountable for cart conversion, order creation, idempotency, and the customer's final purchase step. Give the next shift a defensible finding and make clear which current observations rule out the most tempting alternative explanations. The incident owner is using this finding to set the 2026-03-04 recovery plan for checkout. Turn the finding into a recovery promise the next shift can defend: supported date, schedule variance, cost, approval basis, and the consequence of waiting.\n","reference_tool_calls":35,"source_split":"train","source_task_digest":"84f1c2c8b2197acb3b0e635dca59d0ea2ae388e1c5d426ff3e69a5216e66473a"},{"category":"aiops_analysis","id":"dob100-004-rca-media-disk-pressure","prompt":"Media uploads stalling. The incident record describes the visible effect, but it does not establish whether the apparent owner, a dependency, or the deployed configuration is responsible. Current customer and operator exposure includes customer uploads, image transformation, object storage, and CDN delivery behavior. Establish the operative cause instead of restating the alert, keep production unchanged, and record why the present system state supports that conclusion in the existing incident work item. This diagnosis will decide whether media service enters recovery on 2026-03-04. Compare the staffed recovery path with paid acceleration and a separately approved exception, without borrowing capacity protected for another incident.\n","reference_tool_calls":34,"source_split":"train","source_task_digest":"583a1c5512ebf67af48dafb621918d3c92d9fbca3e762d65d77031752e90d9af"},{"category":"aiops_analysis","id":"dob100-005-rca-notifications-node-deadlock","prompt":"Notifications are not being delivered. The incident record describes the visible effect, but it does not establish whether the apparent owner, a dependency, or the deployed configuration is responsible. The practical blast radius covers receipts, retry policy, delivery templates, and downstream messaging reliability. Separate upstream cause from downstream symptoms and leave a concise, evidence-linked handoff that another engineer can independently challenge. Notifications is tentatively booked for recovery on 2026-03-05; the evidence here determines whether that promise is credible. Before closing the incident record, give the incoming owner the earliest credible recovery date, the constraint behind it, and whether approved acceleration or an exception would materially help.\n","reference_tool_calls":33,"source_split":"train","source_task_digest":"47096a695ba9b655f995c4e1068ff7438a006e23e2524448530847b958bf31fa"},{"category":"aiops_analysis","id":"dob100-006-rca-search-unscheduled-reindex","prompt":"Search results are stale. The incident record describes the visible effect, but it does not establish whether the apparent owner, a dependency, or the deployed configuration is responsible. For search, a credible answer has to reconcile index freshness, query relevance, cache behavior, and shopper discovery latency. Determine what the operating evidence supports today and explain why stale records or plausible shortcuts do not change the answer. The incident owner is using this finding to set the 2026-03-05 recovery plan for search. Say whether the date holds against current on-call commitments and change windows, what a sanctioned acceleration costs, and which faster route still lacks authority.\n","reference_tool_calls":32,"source_split":"heldout","source_task_digest":"2f56f4c60040b7ecf33b76734d8db8b1691e08a4e0e715e03fc16f6559073310"},{"category":"aiops_detection","id":"dob100-007-detect-checkout-errors","prompt":"Is checkout violating its SLOs? The current monitoring views do not yet establish whether this is an active objective breach, a stale signal, or a healthy service being blamed for a downstream symptom. The live operating picture includes cart conversion, order creation, idempotency, and the customer's final purchase step. The disputed mechanism is application exceptions, sampled error groups, request failures, and service-level breach. Trace the inconsistency to its authoritative source and record the supported finding while leaving every production control untouched. This diagnosis will decide whether checkout enters recovery on 2026-03-05. Turn the finding into a recovery promise the next shift can defend: supported date, schedule variance, cost, approval basis, and the consequence of waiting.\n","reference_tool_calls":32,"source_split":"train","source_task_digest":"a39c829d0cc38b5a5f8c0c29b4d95c18141f17096737fe8318f94d802f58b35b"},{"category":"aiops_detection","id":"dob100-008-detect-inventory","prompt":"Determine whether reservation failures are a capacity incident. The current monitoring views do not yet establish whether this is an active objective breach, a stale signal, or a healthy service being blamed for a downstream symptom. For inventory → checkout capability chain, the operational risk spans the upstream-to-downstream dependency direction, the contract at each service boundary, ownership, recoverability, and end-to-end health. Reconcile the affected boundary, identify the decisive signal, and record a production-safe conclusion the service owner can use in the existing incident work item. Checkout is tentatively booked for recovery on 2026-03-05; the evidence here determines whether that promise is credible. Compare the staffed recovery path with paid acceleration and a separately approved exception, without borrowing capacity protected for another incident.\n","reference_tool_calls":38,"source_split":"train","source_task_digest":"6c7d1b71b94b6c84b6c410849e34a2c47701f649f95e0a849549dabaf748eaf6"},{"category":"reconciliation","id":"dob100-009-rcn-checkout-error-rate","prompt":"What is checkout's production error rate? The commerce lead is asking for checkout's current production error rate as a percentage. Sentry and Prometheus give very different numbers. Reconcile the governing definition, time window, identities, and conflicting systems; give the supported answer and explain every material exclusion or unresolved ambiguity. The decision must hold across cart conversion, order creation, idempotency, and the customer's final purchase step. Use the governing definition and present state to make the decision, then leave the shared record consistent for the next workflow. The team plans to publish this answer through checkout on 2026-03-05. State when the result can reliably publish after current refreshes, protected workloads, and review capacity, plus what an approved acceleration would cost.\n","reference_tool_calls":32,"source_split":"train","source_task_digest":"3def2a8fd545a0da09ecb68e8d21456a17c3f3b993ea5d87edc6a187eefd0db6"},{"category":"aiops_detection","id":"dob100-010-detect-status-page-recurrence","prompt":"Customers still cannot check out, and the status page says we fixed it. The current monitoring views do not yet establish whether this is an active objective breach, a stale signal, or a healthy service being blamed for a downstream symptom. The next owner needs this settled across cart conversion, order creation, idempotency, and the customer's final purchase step. The disputed mechanism is a supposedly resolved incident recurring behind stale customer-status communication. Find the narrowest explanation that fits every material signal and leave the incident record ready for independent verification. This diagnosis will decide whether checkout enters recovery on 2026-03-06. Say whether the date holds against current on-call commitments and change windows, what a sanctioned acceleration costs, and which faster route still lacks authority.\n","reference_tool_calls":32,"source_split":"heldout","source_task_digest":"a7e4125418d0c197145370137a7046af75a83e60c2bd55c8bd1eb57583535422"},{"category":"aiops_detection","id":"dob100-011-detect-storefront-healthy","prompt":"Is storefront-web violating its SLOs? The current monitoring views do not yet establish whether this is an active objective breach, a stale signal, or a healthy service being blamed for a downstream symptom. The affected workflow reaches browser delivery, page rendering, customer navigation, and edge-cache behavior. Test the apparent explanation against the live state, distinguish correlation from cause, keep production unchanged, and record the sourced finding in the existing incident work item. Storefront web is tentatively booked for recovery on 2026-03-06; the evidence here determines whether that promise is credible. Turn the finding into a recovery promise the next shift can defend: supported date, schedule variance, cost, approval basis, and the consequence of waiting.\n","reference_tool_calls":32,"source_split":"train","source_task_digest":"d159eccd366b93a655ac4ccb9de77efb49e668513e7218536e95df6e57e51de8"},{"category":"aiops_localization","id":"dob100-012-localize-analytics-crashloop","prompt":"Alarm 9609 is still firing, yet the error tracker is empty and the label may point to a symptom rather than the failing component. Establish the most defensible diagnosis from the current operating evidence, preserve any genuine uncertainty, and record the finding in the existing incident work item without changing production controls. The incident owner is using this finding to set the 2026-03-06 recovery plan for analytics worker. Tell them whether staffed capacity can meet that date, what paid acceleration changes, and which faster option still needs separate approval without taking capacity reserved for another incident.\n","reference_tool_calls":32,"source_split":"train","source_task_digest":"c98cc9d60f057025a4621e3a3bd2e5903ce7991894d14aeddbccbb11fba5ec85"},{"category":"aiops_localization","id":"dob100-013-localize-analytics-errors","prompt":"Alarm 9609 (analytics-worker errors). The label attached to the alarm may describe where the symptom was noticed rather than the component or runtime condition that actually caused it. For analytics worker, the operational risk spans reporting freshness, warehouse replica access, scheduled rollups, and downstream decision data. The disputed mechanism is application exceptions, sampled error groups, request failures, and service-level breach. Resolve the disagreement far enough for the incident lead to act, keep production unchanged, and record both the supported conclusion and any uncertainty the evidence cannot settle. The incident owner is using this finding to set the 2026-03-06 recovery plan for analytics worker. Say whether the date holds against current on-call commitments and change windows, what a sanctioned acceleration costs, and which faster route still lacks authority.\n","reference_tool_calls":32,"source_split":"train","source_task_digest":"a2d23e0eb2d9b8d396a2957c67038ac1a0082f3a042e6d58877e41100efb27b1"},{"category":"aiops_localization","id":"dob100-014-localize-checkout-latency","prompt":"Alarm 9610 (checkout latency). The label attached to the alarm may describe where the symptom was noticed rather than the component or runtime condition that actually caused it. The decision must hold across the upstream-to-downstream dependency direction, the contract at each service boundary, ownership, recoverability, and end-to-end health. Give the next shift a defensible finding and make clear which current observations rule out the most tempting alternative explanations. This diagnosis will decide whether payments enters recovery on 2026-03-06. Turn the finding into a recovery promise the next shift can defend: supported date, schedule variance, cost, approval basis, and the consequence of waiting.\n","reference_tool_calls":33,"source_split":"train","source_task_digest":"a41341181d0b8b97e88b20966bd3cb3a252236f88b870a0bf4eaf86617f0311e"},{"category":"aiops_localization","id":"dob100-015-localize-gateway-latency","prompt":"Alarm 9604 (api-gateway latency). The label attached to the alarm may describe where the symptom was noticed rather than the component or runtime condition that actually caused it. The next owner needs this settled across edge routing, partner request compatibility, authentication policy, and public latency exposure. Establish the operative cause instead of restating the alert, keep production unchanged, and record why the present system state supports that conclusion in the existing incident work item. API gateway is tentatively booked for recovery on 2026-03-07; the evidence here determines whether that promise is credible. Compare the staffed recovery path with paid acceleration and a separately approved exception, without borrowing capacity protected for another incident.\n","reference_tool_calls":32,"source_split":"heldout","source_task_digest":"5769574003ef52c67645d83e4bd9c9542a3ac1350c4d7d5702b12e31c243ddb2"},{"category":"aiops_localization","id":"dob100-016-localize-media-latency","prompt":"Alarm 9607 (media-service latency). The label attached to the alarm may describe where the symptom was noticed rather than the component or runtime condition that actually caused it. The affected workflow reaches customer uploads, image transformation, object storage, and CDN delivery behavior. Separate upstream cause from downstream symptoms and leave a concise, evidence-linked handoff that another engineer can independently challenge. The incident owner is using this finding to set the 2026-03-07 recovery plan for media service. Before closing the incident record, give the incoming owner the earliest credible recovery date, the constraint behind it, and whether approved acceleration or an exception would materially help.\n","reference_tool_calls":32,"source_split":"train","source_task_digest":"80cf22fa7d067fd05c970541214e6c3249e8ec4d7ad62d38e65c680f2e2fd2ec"},{"category":"api_migration","id":"dob100-017-auth-v1-to-v2","prompt":"Migrate /v1/auth traffic to /v2/auth and retire v1. Consumers still span the old and replacement contract, so establish actual usage, compatibility, and rollback conditions before completing the traffic transition. This cannot be closed without accounting for the upstream-to-downstream dependency direction, the contract at each service boundary, ownership, recoverability, and end-to-end health. The change boundary covers tokens, identity sessions, compatibility headers, and authentication consumers. Choose the supported rollout path from the evidence available today, stop if its control conditions are not met, and document what you observed afterward. The supported change is supposed to take API gateway through its 2026-03-07 release window. Give the release owner an honest completion date and show whether the normal path, paid expedite, or a governed exception best protects customers and rollback safety.\n","reference_tool_calls":63,"source_split":"train","source_task_digest":"c213130334d7feeb89732bdf0464da5a850bcee46ebb7857ce0fb9a9ad63e394"},{"category":"aiops_analysis","id":"dob100-018-rca-payments-retry","prompt":"Payments error rate. The incident record describes the visible effect, but it does not establish whether the apparent owner, a dependency, or the deployed configuration is responsible. The decision affects authorization, settlement, retry safety, and duplicate-charge prevention. Trace the inconsistency to its authoritative source and record the supported finding while leaving every production control untouched. Payments is tentatively booked for recovery on 2026-03-07; the evidence here determines whether that promise is credible. Turn the finding into a recovery promise the next shift can defend: supported date, schedule variance, cost, approval basis, and the consequence of waiting.\n","reference_tool_calls":33,"source_split":"train","source_task_digest":"05fcf362effa2357689b567acf74bbc9a25deed150a2a2b1bb6f6bdce5ba022a"},{"category":"aiops_analysis","id":"dob100-019-rca-catalog-n-plus-one","prompt":"Catalog pricing latency. The incident record describes the visible effect, but it does not establish whether the apparent owner, a dependency, or the deployed configuration is responsible. The handoff has to remain sound across price publication, product availability, merchandising feeds, and cache-consistent customer display. Reconcile the affected boundary, identify the decisive signal, and record a production-safe conclusion the service owner can use in the existing incident work item. The incident owner is using this finding to set the 2026-03-07 recovery plan for catalog. Compare the staffed recovery path with paid acceleration and a separately approved exception, without borrowing capacity protected for another incident.\n","reference_tool_calls":34,"source_split":"train","source_task_digest":"1abbb709997b1cf2090e37e25f93674c9cc95e1aae9e5c0f9abe6724e4e6d228"},{"category":"aiops_analysis","id":"dob100-020-rca-notifications-timeout","prompt":"Notification delivery failures. The incident record describes the visible effect, but it does not establish whether the apparent owner, a dependency, or the deployed configuration is responsible. The service owner is accountable for receipts, retry policy, delivery templates, and downstream messaging reliability. The disputed mechanism is deadline propagation, downstream waiting, retry amplification, and request cancellation. Work through the conflicting records, preserve genuine uncertainty, and state the supported operational conclusion in ordinary language. This diagnosis will decide whether notifications enters recovery on 2026-03-08. Before closing the incident record, give the incoming owner the earliest credible recovery date, the constraint behind it, and whether approved acceleration or an exception would materially help.\n","reference_tool_calls":34,"source_split":"train","source_task_digest":"321cf65045e0fcc0fbe5fc61987e57bb3f762c718bdf961faff70c27b931b8a1"},{"category":"aiops_analysis","id":"dob100-021-rca-inventory-pool","prompt":"Inventory reservation failures. The incident record describes the visible effect, but it does not establish whether the apparent owner, a dependency, or the deployed configuration is responsible. Current customer and operator exposure includes stock reservation, warehouse truth, backorder promises, and race-free quantity updates. Find the narrowest explanation that fits every material signal and leave the incident record ready for independent verification. Inventory is tentatively booked for recovery on 2026-03-08; the evidence here determines whether that promise is credible. Say whether the date holds against current on-call commitments and change windows, what a sanctioned acceleration costs, and which faster route still lacks authority.\n","reference_tool_calls":32,"source_split":"train","source_task_digest":"120f2833653df68458e0db8c3df3fddfe4bee0d9b28fcf298673c6520a09dcc9"},{"category":"aiops_analysis","id":"dob100-022-rca-inventory-migrator-security-context","prompt":"Inventory counts disagree with the database. The incident record describes the visible effect, but it does not establish whether the apparent owner, a dependency, or the deployed configuration is responsible. The practical blast radius covers stock reservation, warehouse truth, backorder promises, and race-free quantity updates. Test the apparent explanation against the live state, distinguish correlation from cause, keep production unchanged, and record the sourced finding in the existing incident work item. The incident owner is using this finding to set the 2026-03-08 recovery plan for inventory. Turn the finding into a recovery promise the next shift can defend: supported date, schedule variance, cost, approval basis, and the consequence of waiting.\n","reference_tool_calls":35,"source_split":"train","source_task_digest":"2635fdc9f20ff7efc3d516cf181a75ac9df210ea236663e343f9f316d32fb6e7"},{"category":"aiops_analysis","id":"dob100-023-rca-inventory-missing-role","prompt":"The inventory report has never produced output, and the incident record does not establish whether the owner, a dependency, or deployed configuration is responsible. Settle the root cause as far as the live evidence allows and record the supported finding in the incident work item, keeping production unchanged and genuine uncertainties visible. This diagnosis will decide whether inventory enters recovery on 2026-03-08. Give the next shift a defensible date, the binding capacity or change-window constraint, the cost of sanctioned acceleration, and the exception route that remains unapproved without displacing another team's reservation.\n","reference_tool_calls":34,"source_split":"heldout","source_task_digest":"0b91d55a88bdc20cdd526d7075f544c877186e330be91ee3c4f41f3d5ec9886a"},{"category":"attribution","id":"dob100-024-attr-three-at-once","prompt":"Three alarms in twenty minutes - is this one incident or three? Three customer-impacting alarms arrived within the same twenty-minute window. Determine whether one mechanism explains them or whether they are independent failures, localize each supported cause, and record a production-safe finding that tells the incident lead how many investigations are actually needed. The live operating picture includes temporal correlation versus shared causation, the distinct runtime and deployment evidence for each affected service, and whether one or several incident owners are required. Explain which conclusion survives cross-checking, record it in the existing incident work item, and leave the live evidence that lets the incoming engineer reproduce it; production controls should remain unchanged. Checkout is tentatively booked for recovery on 2026-03-08; the evidence here determines whether that promise is credible. Before closing the incident record, give the incoming owner the earliest credible recovery date, the constraint behind it, and whether approved acceleration or an exception would materially help.\n","reference_tool_calls":38,"source_split":"train","source_task_digest":"c17f7c82251a25f377444772fdb1953cec5a4232cdbf9f66df615865572a790a"},{"category":"aiops_analysis","id":"dob100-025-rca-catalog-gc-thrash","prompt":"Catalog latency comes in waves. The incident record describes the visible effect, but it does not establish whether the apparent owner, a dependency, or the deployed configuration is responsible. The service owner is accountable for price publication, product availability, merchandising feeds, and cache-consistent customer display. Give the next shift a defensible finding and make clear which current observations rule out the most tempting alternative explanations. Catalog is tentatively booked for recovery on 2026-03-09; the evidence here determines whether that promise is credible. Turn the finding into a recovery promise the next shift can defend: supported date, schedule variance, cost, approval basis, and the consequence of waiting.\n","reference_tool_calls":32,"source_split":"heldout","source_task_digest":"578b32c08a4cf1a459afb573a37c1a7f06aa48ed94c701a4feb7f1cfe096ee13"},{"category":"aiops_analysis","id":"dob100-026-rca-storefront-traffic-flood","prompt":"The homepage is taking unusual load. The incident record describes the visible effect, but it does not establish whether the apparent owner, a dependency, or the deployed configuration is responsible. Current customer and operator exposure includes browser delivery, page rendering, customer navigation, and edge-cache behavior. Establish the operative cause instead of restating the alert, keep production unchanged, and record why the present system state supports that conclusion in the existing incident work item. The incident owner is using this finding to set the 2026-03-09 recovery plan for storefront web. Compare the staffed recovery path with paid acceleration and a separately approved exception, without borrowing capacity protected for another incident.\n","reference_tool_calls":34,"source_split":"train","source_task_digest":"4f1eac0dbf2f78e657c08f47210dbe866e2e07ad48a5cfa972fc512e8ac6e41b"},{"category":"aiops_analysis","id":"dob100-027-rca-analytics-untolerated-taint","prompt":"The nightly reconciliation never runs. The incident record describes the visible effect, but it does not establish whether the apparent owner, a dependency, or the deployed configuration is responsible. The practical blast radius covers reporting freshness, warehouse replica access, scheduled rollups, and downstream decision data. Separate upstream cause from downstream symptoms and leave a concise, evidence-linked handoff that another engineer can independently challenge. This diagnosis will decide whether analytics worker enters recovery on 2026-03-09. Before closing the incident record, give the incoming owner the earliest credible recovery date, the constraint behind it, and whether approved acceleration or an exception would materially help.\n","reference_tool_calls":34,"source_split":"train","source_task_digest":"a1a8c64d99bfa53d2c0f93b7c4d6337d43c9e46331b1540bb9a867b14629937c"},{"category":"aiops_analysis","id":"dob100-028-rca-media-recreate-strategy","prompt":"Media-service is briefly unavailable on every release. The incident record describes the visible effect, but it does not establish whether the apparent owner, a dependency, or the deployed configuration is responsible. For media service, a credible answer has to reconcile customer uploads, image transformation, object storage, and CDN delivery behavior. Determine what the operating evidence supports today and explain why stale records or plausible shortcuts do not change the answer. Media service is tentatively booked for recovery on 2026-03-09; the evidence here determines whether that promise is credible. Say whether the date holds against current on-call commitments and change windows, what a sanctioned acceleration costs, and which faster route still lacks authority.\n","reference_tool_calls":36,"source_split":"heldout","source_task_digest":"a327f07023a8dc847d15dac055f37bbe9b5ae923a6b77b624d4b731f8f341f22"},{"category":"aiops_analysis","id":"dob100-029-rca-inventory-unbound-storage","prompt":"Inventory is one replica short. The incident record describes the visible effect, but it does not establish whether the apparent owner, a dependency, or the deployed configuration is responsible. The live operating picture includes stock reservation, warehouse truth, backorder promises, and race-free quantity updates. Trace the inconsistency to its authoritative source and record the supported finding while leaving every production control untouched. The incident owner is using this finding to set the 2026-03-09 recovery plan for inventory. Turn the finding into a recovery promise the next shift can defend: supported date, schedule variance, cost, approval basis, and the consequence of waiting.\n","reference_tool_calls":34,"source_split":"train","source_task_digest":"860bccdbc60e22d39f9746bf7f6e08e85102f0f9c40ad417472cf6a61c972be9"},{"category":"reconciliation","id":"dob100-030-rcn-distinct-checkout-bugs","prompt":"How many distinct open checkout bugs do we have? The commerce lead wants to know how many genuinely distinct open checkout bugs exist. Bugs get filed in more than one tracker. Reconcile the governing definition, time window, identities, and conflicting systems; give the supported answer and explain every material exclusion or unresolved ambiguity. For checkout, the operational risk spans cart conversion, order creation, idempotency, and the customer's final purchase step. Follow the evidence through its conflicting representations, resolve the employee's underlying question, and make the conclusion reproducible. The reconciled answer is due in the 2026-03-04 operating review for checkout. Record whether the normal cycle meets the window, the constraint that governs it, and the consequences of waiting, expediting, or seeking broader authority.\n","reference_tool_calls":32,"source_split":"heldout","source_task_digest":"8d1e63ef730686ae64ba9031fce55fb2fb2986d5bdae39b16a3921494d62fb91"},{"category":"code_implementation","id":"dob100-031-impl-backoff","prompt":"Implement the payments retry backoff schedule. The repository documents the intended behavior but the executable boundary is incomplete; infer the non-obvious cases from its consumers and leave the behavior demonstrably correct. The decision must hold across authorization, settlement, retry safety, and duplicate-charge prevention. The implementation boundary is retry delay growth, caps, jitter behavior, and transient-failure safety. Implement the supported behavior with a minimal diff, verify its failure modes as well as its happy path, and document any remaining assumption. Before the engineering ticket closes, decide whether payments retry worker can carry the fix into the 2026-03-04 window. Include where the fix fits in the release train, whether reviewer and pipeline capacity support the date, and what paid acceleration or an exception would actually change.\n","reference_tool_calls":33,"source_split":"train","source_task_digest":"38da2f8ec128370470dfa262acac43b809383c3bdd733bdc392958e2b8e6475a"},{"category":"code_implementation","id":"dob100-032-impl-cachekey","prompt":"Implement the search cache key. The repository documents the intended behavior but the executable boundary is incomplete; infer the non-obvious cases from its consumers and leave the behavior demonstrably correct. The next owner needs this settled across index freshness, query relevance, cache behavior, and shopper discovery latency. Reconcile the written contract with the executable behavior, correct the discrepancy, and make the result straightforward for another engineer to reproduce. The team wants this change on the search cache worker release train for 2026-03-04. Give the maintainer a credible completion rather than the target milestone: account for protected engineering work, release windows, cost, and approval authority.\n","reference_tool_calls":33,"source_split":"train","source_task_digest":"19472d2f8b041f56a0b66123f5d0547b2b518e12389aaf8fa1790470387de033"},{"category":"code_implementation","id":"dob100-033-impl-chunk","prompt":"Implement settlement batch chunking. The repository documents the intended behavior but the executable boundary is incomplete; infer the non-obvious cases from its consumers and leave the behavior demonstrably correct. The affected workflow reaches authorization, settlement, retry safety, and duplicate-charge prevention. The implementation boundary is batch partitioning, payload size, terminal fragments, and ordering. Use the current workspace evidence to identify the missing invariant, restore it without masking adjacent failures, and prove the final behavior locally. The repaired behavior is intended for the 2026-03-04 release of payments settlement worker. Compare the normal release path with an approved expedite and the faster exception path, then record the earliest date the evidence truly supports.\n","reference_tool_calls":33,"source_split":"train","source_task_digest":"709db641d6b7afdfa11bd31e5d4a8a0d5508c95585fcee5228ba214c353b714e"},{"category":"code_implementation","id":"dob100-034-impl-ratelimit","prompt":"Implement per-client rate limiting at the edge. The repository documents the intended behavior but the executable boundary is incomplete; infer the non-obvious cases from its consumers and leave the behavior demonstrably correct. This cannot be closed without accounting for edge routing, partner request compatibility, authentication policy, and public latency exposure. Close the engineering request at the behavior level, not merely the visible assertion, and leave a concise explanation of what changed and why. Before the engineering ticket closes, decide whether API rate limit service can carry the fix into the 2026-03-04 window. State whether the requested window survives the current queue, what constraint controls it, and the cost and authority trade-off of improving the date.\n","reference_tool_calls":33,"source_split":"heldout","source_task_digest":"815a47f9482c175d1bd987b8dd9ea72836498356b2246bf05efd55fcedf2ee0b"},{"category":"cross_system","id":"dob100-035-port-close-backlog-issues","prompt":"Empty the Backlog before sprint planning. Sprint planning wants the Backlog emptied: move every issue currently in Backlog to Done with a resolution recording why. Issues in any other state are somebody's work in flight and must not be touched. The decision affects tracker identity, status, ownership, duplicate control, and communication to the teams acting on the resulting population. Use the current records to decide what genuinely belongs in scope, record the supported answer in the existing work item, and leave a result that can be independently audited. The cross-system handoff is scheduled through engineering portfolio automation for 2026-03-05. Give the requesting team a supported delivery date based on the real queue and protected commitments, including the cost and authority needed to improve it.\n","reference_tool_calls":35,"source_split":"train","source_task_digest":"ad08e6dd271c7361d2024a30fe9a92182648fd21e069899fa3c5f91c3b550ea1"},{"category":"cross_system","id":"dob100-036-port-close-blocked-issues","prompt":"Clear the Blocked column. Anything sitting in Blocked has been superseded by work already in flight. Move every Blocked issue to Done, recording a resolution that says so. Nothing else moves - In Progress issues are live and Backlog is next sprint's problem. The handoff has to remain sound across tracker identity, status, ownership, duplicate control, and communication to the teams acting on the resulting population. Resolve the cross-system disagreement before acting, protect unrelated work, and confirm the final record from its authoritative source. The requesting team expects the result through engineering portfolio automation by 2026-03-05. Compare the normal handoff, approved acceleration, and exception path on timing and operational consequence without displacing another team's reservation.\n","reference_tool_calls":33,"source_split":"train","source_task_digest":"12abe32f8249b9ca354b30910d927b6efbab1de1ad05a2fd9fb9fc55e8103cff"},{"category":"cross_system","id":"dob100-037-port-copy-priority-issues","prompt":"Mirror priority GitHub issues onto the board. Every OPEN issue carrying the 'priority' label needs a matching ticket in our own tracker so it appears on the board. Create one ticket per such issue, citing the issue number it came from in the description. Issues that are closed, or that lack the label, must not be copied. Tell #eng once they are mirrored, so nobody picks one up twice. The next owner needs this settled across tracker identity, status, ownership, duplicate control, and communication to the teams acting on the resulting population. Treat duplicate, stale, and in-flight records carefully, make only the scoped update, and verify that the intended population changed. The requesting team expects the result through engineering portfolio automation by 2026-03-05. Leave one independently checkable recommendation with its completion date, incremental cost, approval basis, binding constraint, and rejected alternative.\n","reference_tool_calls":40,"source_split":"heldout","source_task_digest":"f80732566d0d8b84950a68ec9d36c7a78451c2458b1828a141936be1bb433189"},{"category":"cross_system","id":"dob100-038-port-count-inflight-work","prompt":"How much high-priority work is in flight? How many issues are In Progress at High priority or above? Highest counts as above High. Give the requesting team the supported result and cite the governing records. The affected workflow reaches tracker identity, status, ownership, duplicate control, and communication to the teams acting on the resulting population. Reconcile identity across the available systems, apply the business rule to the resulting set, and leave a clear account of exclusions. Engineering portfolio automation owns the 2026-03-05 delivery promise for this request. Give the requesting team a supported delivery date based on the real queue and protected commitments, including the cost and authority needed to improve it.\n","reference_tool_calls":32,"source_split":"train","source_task_digest":"e4482fd7aa4e749dcd513faa98a4bef3eb79c0a1cb4d3cfcc13a5d38340e028e"},{"category":"cross_system","id":"dob100-039-port-count-unowned-work","prompt":"How much unassigned work is still live? Planning wants to know how much live work has nobody on it: how many tickets are unassigned and not already finished? Give the requesting team the supported result and cite the governing records., and state in your assumptions how you treated finished tickets - an unassigned ticket that is already done needs no owner, and counting it overstates the problem. This cannot be closed without accounting for tracker identity, status, ownership, duplicate control, and communication to the teams acting on the resulting population. Establish which source governs the decision, complete the requested coordination, and make any unresolved exception visible to its owner. The cross-system handoff is scheduled through engineering portfolio automation for 2026-03-05. Compare the normal handoff, approved acceleration, and exception path on timing and operational consequence without displacing another team's reservation.\n","reference_tool_calls":32,"source_split":"heldout","source_task_digest":"50389b3dc9960fb7858e12a98acfcf35a3ed11044553f8569a686877d1dca599"},{"category":"cross_system","id":"dob100-040-port-dedupe-linked-issues","prompt":"Close the tickets that duplicate other trackers. Some tickets on our board are duplicates of work already tracked in another system, and we keep re-triaging them. Close every ticket that is recorded as duplicating an issue in another tracker, with a resolution that says so. A ticket that merely looks similar is not a duplicate - only the recorded links count, and anything without one stays exactly as it is. The decision affects tracker identity, status, ownership, duplicate control, and communication to the teams acting on the resulting population. Work from the actual current population rather than a convenient snapshot, preserve out-of-scope records, and verify the delivered outcome. The requesting team expects the result through engineering portfolio automation by 2026-03-06. If the promised date is not credible, record the earliest workable alternative, its schedule variance, and the approval still needed for anything faster.\n","reference_tool_calls":35,"source_split":"train","source_task_digest":"6b05c38c6785b92b4a9e170aa3d5f5d69f444025e9f01e3ebbdfbcd386ea6682"},{"category":"cross_system","id":"dob100-041-port-escalate-to-gateway-oncall","prompt":"Who do we page for the gateway right now? The gateway is in trouble and we are about to page. Name the person who is actually on call for it right now, Give the requesting team the supported result and cite the governing records. Paging the wrong person costs us the incident, so go through the escalation policy that covers the gateway rather than whoever is most visible on the schedule - and note that the paging system does not call the gateway what we call it. The handoff has to remain sound across edge routing, partner request compatibility, authentication policy, and public latency exposure. Follow the evidence through its conflicting representations, resolve the employee's underlying question, and make the conclusion reproducible. API gateway owns the 2026-03-06 delivery promise for this request. Leave one independently checkable recommendation with its completion date, incremental cost, approval basis, binding constraint, and rejected alternative.\n","reference_tool_calls":32,"source_split":"train","source_task_digest":"888eb1b723c70a4463a1d378454feb01329b54542f1271f9d9dcb3e72799842d"},{"category":"cross_system","id":"dob100-042-port-high-priority-since","prompt":"Chase the recent high-priority issues in #eng. Post one message to #eng naming every issue that is OPEN, carries the 'priority' label, AND was created on day 410 or later. Reference each by number and ask whether anyone is still on it. Issues that miss any one of those three conditions must not appear - the point of the message is that everything in it needs an owner today. The service owner is accountable for tracker identity, status, ownership, duplicate control, and communication to the teams acting on the resulting population. Use the governing definition and present state to make the decision, then leave the shared record consistent for the next workflow. The cross-system handoff is scheduled through engineering portfolio automation for 2026-03-06. Give the requesting team a supported delivery date based on the real queue and protected commitments, including the cost and authority needed to improve it.\n","reference_tool_calls":33,"source_split":"train","source_task_digest":"013f4b3437944c21ac7c757d4590677290737ca313e092e372ad716f1c511838"},{"category":"error_rate_reduction","id":"dob100-043-analytics-batch-size","prompt":"Analytics rollups run in batches large enough to amplify memory pressure. Large rollup batches amplify memory pressure on the consumer. Determine which current mechanism actually explains the degradation, choose the smallest supported production repair, and verify that the customer-facing objective recovers. Current customer and operator exposure includes reporting freshness, warehouse replica access, scheduled rollups, and downstream decision data. Complete the production outcome without broadening scope, confirm the relevant health signal, and communicate any exception that still needs ownership. The supported change is supposed to take analytics worker through its 2026-03-06 release window. Give the release owner an honest completion date and show whether the normal path, paid expedite, or a governed exception best protects customers and rollback safety.\n","reference_tool_calls":40,"source_split":"train","source_task_digest":"779a20083e0b55767175b636f4ce7ad536f523bee37686451e2842956e49d1bb"},{"category":"error_rate_reduction","id":"dob100-044-analytics-prefetch","prompt":"Analytics worker restarting under queue load. Analytics-worker error_rate_pct is 6.0% against a 2.0% SLO (alarm 9609). Determine which current mechanism actually explains the degradation, choose the smallest supported production repair, and verify that the customer-facing objective recovers. The practical blast radius covers reporting freshness, warehouse replica access, scheduled rollups, and downstream decision data. Use current operating constraints to select the viable option, execute its guarded transition, and demonstrate the post-change state at the service boundary. Customers are expecting the result from analytics worker in the 2026-03-06 window, so the change is not complete until that promise is credible. If the committed window will slip, identify the binding constraint and the earliest defensible alternative rather than taking another team's reserved slot.\n","reference_tool_calls":48,"source_split":"train","source_task_digest":"c12233a43aeb590a928066907dd0af0b78f3ced2637a07b9d566ae799bbe85d4"},{"category":"error_rate_reduction","id":"dob100-045-checkout-payments-timeout","prompt":"Checkout waits 8s on the payments call. The current signal is beyond the standard. Determine which current mechanism actually explains the degradation, choose the smallest supported production repair, and verify that the customer-facing objective recovers. For checkout → payments capability chain, a credible answer has to reconcile the upstream-to-downstream dependency direction, the contract at each service boundary, ownership, recoverability, and end-to-end health. The disputed mechanism is deadline propagation, downstream waiting, retry amplification, and request cancellation. Balance restoration speed with dependency and rollback risk, then leave a sourced handoff showing why the final production state is acceptable. Operations needs one go-or-no-go call for the change and the 2026-03-07 cutover of payments. Record the chosen rollout date, schedule variance, incremental cost, and approval basis, along with the tempting option the current authority does not support.\n","reference_tool_calls":45,"source_split":"train","source_task_digest":"7f1902b240e634e8fb5da9357cc23b837c05ab730ce04366c5d7f0850a7563ae"},{"category":"error_rate_reduction","id":"dob100-046-inventory-pool","prompt":"Inventory reservations are failing under peak traffic: error rate is 4.7% against a 1.0% SLO, alarm 9606 is firing, and incident 9703 remains open. Restore the customer-facing objective with the smallest production repair the current evidence and approval actually support, and leave unrelated service state untouched. The supported change is supposed to take inventory through its 2026-03-07 release window. Give the release owner an honest go-or-no-go with the earliest defensible completion, the binding constraint and acceleration cost, and any faster option that still needs separate authority rather than another team's reserved capacity.\n","reference_tool_calls":50,"source_split":"heldout","source_task_digest":"73485fe16c960c5297072b2e4d8caf84362bdd3e1b20c5ed5b1c82bf30f8b761"},{"category":"error_rate_reduction","id":"dob100-047-notifications-timeout","prompt":"Notification delivery failures from hung SMTP calls. Notifications error_rate_pct is 3.6% against a 1.5% SLO (alarm 9608). Determine which current mechanism actually explains the degradation, choose the smallest supported production repair, and verify that the customer-facing objective recovers. For notifications, the operational risk spans receipts, retry policy, delivery templates, and downstream messaging reliability. The disputed mechanism is deadline propagation, downstream waiting, retry amplification, and request cancellation. Carry the remediation through only where live conditions permit it, then reopen the affected state to prove the intended outcome actually holds. Customers are expecting the result from notifications in the 2026-03-07 window, so the change is not complete until that promise is credible. Give the release owner an honest completion date and show whether the normal path, paid expedite, or a governed exception best protects customers and rollback safety.\n","reference_tool_calls":47,"source_split":"train","source_task_digest":"47a2b14f025aef1057b5af07d733e21064fa097bbd6922fc0ca6470bfd545892"},{"category":"cross_system","id":"dob100-048-port-count-open-priority","prompt":"How many open priority issues are there? The weekly report needs one number: how many issues are OPEN and labelled 'priority' right now? Give the requesting team the supported result and cite the governing records. The decision must hold across tracker identity, status, ownership, duplicate control, and communication to the teams acting on the resulting population. Determine the supported answer or transition from the live evidence, then communicate it at the level the requesting team needs. The cross-system handoff is scheduled through engineering portfolio automation for 2026-03-07. If the promised date is not credible, record the earliest workable alternative, its schedule variance, and the approval still needed for anything faster.\n","reference_tool_calls":32,"source_split":"train","source_task_digest":"7e6c14437af18f303b2538d9552713b48f82cea64c05e91f6590a548db3e9419"},{"category":"error_rate_reduction","id":"dob100-049-payments-retry","prompt":"Payments error rate breaching the 1% SLO. Payments error_rate_pct is 4.2% against a 1.0% SLO (alarm 9601). Determine which current mechanism actually explains the degradation, choose the smallest supported production repair, and verify that the customer-facing objective recovers. The practical blast radius covers authorization, settlement, retry safety, and duplicate-charge prevention. Protect unrelated production state, handle the material dependency or migration constraint, and verify the result rather than trusting the change record. Operations needs one go-or-no-go call for the change and the 2026-03-07 cutover of payments. Make the go-or-no-go call against the real change window and team availability; state the supported date, the cost of acceleration, and any exception that remains unapproved.\n","reference_tool_calls":48,"source_split":"train","source_task_digest":"722de78a81e1ece94303a0c64552542589cfbfacb4eb8e46ef1856e766da4c64"},{"category":"error_rate_reduction","id":"dob100-050-payments-retry-from-code","prompt":"Payments: eliminate the permanent-failure path on notification timeouts. Payments error_rate_pct is 4.2% (SLO 1.0%) and the error tracker shows 18k events on 'pay-timeout-01'. Determine which current mechanism actually explains the degradation, choose the smallest supported production repair, and verify that the customer-facing objective recovers. For payments, a credible answer has to reconcile authorization, settlement, retry safety, and duplicate-charge prevention. Choose the supported rollout path from the evidence available today, stop if its control conditions are not met, and document what you observed afterward. The supported change is supposed to take payments through its 2026-03-08 release window. Give the release owner an honest completion date and show whether the normal path, paid expedite, or a governed exception best protects customers and rollback safety.\n","reference_tool_calls":46,"source_split":"heldout","source_task_digest":"7265e01eeac0e43f0357618f6a8e244f165c39942cc088b54e1bff15ff0d19d1"},{"category":"feature_flag","id":"dob100-051-checkout-v2-layout-cleanup","prompt":"Checkout_v2_layout has been fully rolled out for months. The present exposure and customer impact have to be established from the live rollout record before deciding whether this is containment or permanent cleanup. The live operating picture includes cart conversion, order creation, idempotency, and the customer's final purchase step. Resolve the customer-impacting condition with the smallest defensible intervention and prove both the recovery and the retained safety boundary. Customers are expecting the result from checkout in the 2026-03-08 window, so the change is not complete until that promise is credible. If the committed window will slip, identify the binding constraint and the earliest defensible alternative rather than taking another team's reserved slot.\n","reference_tool_calls":43,"source_split":"train","source_task_digest":"5fc3e7d4b2675ac93649414174db16ea75eb8b6f990bf0aaab2c89e5261bc019"},{"category":"feature_flag","id":"dob100-052-express-checkout","prompt":"Ship express checkout behind a feature flag at 10%. The product team has approved a guarded first release, but implementation readiness, ownership, prerequisites, and safe initial exposure still have to be established from the current records. For checkout, the operational risk spans cart conversion, order creation, idempotency, and the customer's final purchase step. The product boundary is accelerated checkout, cart conversion, rollout cohorts, and purchase completion. Account for affected consumers and current authorization, carry the change through in a recoverable way, and leave enough evidence for independent review. Operations needs one go-or-no-go call for the change and the 2026-03-08 cutover of checkout. Record the chosen rollout date, schedule variance, incremental cost, and approval basis, along with the tempting option the current authority does not support.\n","reference_tool_calls":44,"source_split":"train","source_task_digest":"a174f6e8a72b23dee83d2f769f7d9200941a565750fc5add005d56e19b62d95c"},{"category":"feature_flag","id":"dob100-053-instant-refunds-killswitch","prompt":"Checkout error spike since the instant_refunds ramp. The present exposure and customer impact have to be established from the live rollout record before deciding whether this is containment or permanent cleanup. The decision must hold across cart conversion, order creation, idempotency, and the customer's final purchase step. Treat conflicting records as part of the work: establish precedence, make only the justified change, and verify the authoritative state after writing it. The supported change is supposed to take checkout through its 2026-03-08 release window. Make the go-or-no-go call against the real change window and team availability; state the supported date, the cost of acceleration, and any exception that remains unapproved.\n","reference_tool_calls":43,"source_split":"train","source_task_digest":"9fa8002e221f078e46a4db992038a4b6c3901aca578d7a7d57f157a488e86f13"},{"category":"feature_flag","id":"dob100-054-legacy-price-rounding-cleanup","prompt":"Legacy_price_rounding has been fully rolled out for months. The present exposure and customer impact have to be established from the live rollout record before deciding whether this is containment or permanent cleanup. The next owner needs this settled across price publication, product availability, merchandising feeds, and cache-consistent customer display. Complete the production outcome without broadening scope, confirm the relevant health signal, and communicate any exception that still needs ownership. Customers are expecting the result from catalog in the 2026-03-08 window, so the change is not complete until that promise is credible. Give the release owner an honest completion date and show whether the normal path, paid expedite, or a governed exception best protects customers and rollback safety.\n","reference_tool_calls":41,"source_split":"heldout","source_task_digest":"dea70227492c3d9f68af28906926c38bc55708af7377efad6ca394b9a33c7ba6"},{"category":"feature_flag","id":"dob100-055-media-webp","prompt":"Ship WebP delivery behind a feature flag. The product team has approved a guarded first release, but implementation readiness, ownership, prerequisites, and safe initial exposure still have to be established from the current records. The affected workflow reaches customer uploads, image transformation, object storage, and CDN delivery behavior. Use current operating constraints to select the viable option, execute its guarded transition, and demonstrate the post-change state at the service boundary. Operations needs one go-or-no-go call for the change and the 2026-03-09 cutover of media service. If the committed window will slip, identify the binding constraint and the earliest defensible alternative rather than taking another team's reserved slot.\n","reference_tool_calls":41,"source_split":"train","source_task_digest":"84438492b2c21d3e7789e44b8c8ce9311b484d26ab7e0d4cb0ff313463c70bc5"},{"category":"feature_flag","id":"dob100-056-saved-carts","prompt":"Ship saved carts (schema change) behind a feature flag. The product team has approved a guarded first release, but implementation readiness, ownership, prerequisites, and safe initial exposure still have to be established from the current records. This cannot be closed without accounting for cart conversion, order creation, idempotency, and the customer's final purchase step. The product boundary is returning-shopper persistence, cart restoration, ownership, and session continuity. Balance restoration speed with dependency and rollback risk, then leave a sourced handoff showing why the final production state is acceptable. The supported change is supposed to take checkout through its 2026-03-09 release window. Record the chosen rollout date, schedule variance, incremental cost, and approval basis, along with the tempting option the current authority does not support.\n","reference_tool_calls":46,"source_split":"train","source_task_digest":"012234571c2727de7e19ffd262c19f7739c24566ee0d4f1ea7b50a493ef212d9"},{"category":"feature_flag","id":"dob100-057-search-autocomplete","prompt":"The product team wants search autocomplete launched behind a guarded flag, but ownership, prerequisites, implementation readiness, and safe exposure are scattered across current records. Decide whether a 10% production start is supportable and, if it is within recorded approval, apply only the supported change without touching unrelated flags. Customers are expecting the result from search in the 2026-03-09 window, so the change is not complete until that promise is credible. Give the release owner the earliest defensible date, its cost and schedule variance, the binding constraint, and any faster route that still needs separate authority.\n","reference_tool_calls":43,"source_split":"train","source_task_digest":"067a54a363f58bb1dac22ba30fe96677c92c074811b31adc61f00909954da32a"},{"category":"flaky_test","id":"dob100-058-flaky-analytics-rollup","prompt":"Fix flaky test_rollup_window. The team has been rerunning the pipeline until it passes. Determine the real source of nondeterminism and make the affected behavior trustworthy rather than masking the intermittent failure. The handoff has to remain sound across reporting freshness, warehouse replica access, scheduled rollups, and downstream decision data. Reproduce the failure, repair its underlying contract rather than its surface symptom, and leave the available validation trustworthy for the next contributor. The team wants this change on the analytics worker release train for 2026-03-09. Give the maintainer a credible completion rather than the target milestone: account for protected engineering work, release windows, cost, and approval authority.\n","reference_tool_calls":40,"source_split":"train","source_task_digest":"f47b7a25c8924a92031e59ff74311706198cdb863b15e2f4ba8a19b447f9f01d"},{"category":"flaky_test","id":"dob100-059-flaky-catalog-rounding","prompt":"Fix flaky test_price_rounding. The team has been rerunning the pipeline until it passes. Determine the real source of nondeterminism and make the affected behavior trustworthy rather than masking the intermittent failure. The service owner is accountable for price publication, product availability, merchandising feeds, and cache-consistent customer display. Resolve the implementation gap without coupling unrelated behavior, demonstrate the expected edge cases, and record the reasoning behind the change. The repaired behavior is intended for the 2026-03-09 release of catalog. Compare the normal release path with an approved expedite and the faster exception path, then record the earliest date the evidence truly supports.\n","reference_tool_calls":40,"source_split":"train","source_task_digest":"bc6b0728e00bf5ae859a0ce78d16aea6b1b4b9f08ebe442f9557350d04b1f75f"},{"category":"flaky_test","id":"dob100-060-flaky-checkout-idempotency","prompt":"Fix flaky test_checkout_idempotency. The team has been rerunning the pipeline until it passes. Determine the real source of nondeterminism and make the affected behavior trustworthy rather than masking the intermittent failure. Current customer and operator exposure includes cart conversion, order creation, idempotency, and the customer's final purchase step. Treat the existing code and tests as evidence rather than a recipe, close the behavioral gap, and show that the repaired result remains stable under repetition. Before the engineering ticket closes, decide whether checkout can carry the fix into the 2026-03-04 window. State whether the requested window survives the current queue, what constraint controls it, and the cost and authority trade-off of improving the date.\n","reference_tool_calls":40,"source_split":"train","source_task_digest":"fbd718363db1aa70817415a7231435b406d65582f6140f2fb97662d501fa439c"},{"category":"flaky_test","id":"dob100-061-flaky-gateway-timeout","prompt":"Fix flaky test_upstream_timeout. The team has been rerunning the pipeline until it passes. Determine the real source of nondeterminism and make the affected behavior trustworthy rather than masking the intermittent failure. This cannot be closed without accounting for edge routing, partner request compatibility, authentication policy, and public latency exposure. The disputed mechanism is deadline propagation, downstream waiting, retry amplification, and request cancellation. Make the implementation match its consumer-facing contract, cover the non-obvious cases, and keep the handoff clear enough for independent review. Before the engineering ticket closes, decide whether API gateway can carry the fix into the 2026-03-04 window. Give the maintainer a credible completion rather than the target milestone: account for protected engineering work, release windows, cost, and approval authority.\n","reference_tool_calls":40,"source_split":"heldout","source_task_digest":"3977b488ef33460db0bbcba0ab2e359b96ea9f18acfd94c1cbc609846ccbe81b"},{"category":"cross_system","id":"dob100-062-w6-copy-13","prompt":"Mirror the issues that are open and raised on day 404 or later. Every issue that is open and raised on day 404 or later needs a matching ticket on our board, citing the issue number it came from. Nothing outside the filter gets copied. Tell #eng once they are mirrored. The decision affects tracker identity, status, ownership, duplicate control, and communication to the teams acting on the resulting population. Work from the actual current population rather than a convenient snapshot, preserve out-of-scope records, and verify the delivered outcome. The cross-system handoff is scheduled through engineering portfolio automation for 2026-03-04. If the promised date is not credible, record the earliest workable alternative, its schedule variance, and the approval still needed for anything faster.\n","reference_tool_calls":48,"source_split":"train","source_task_digest":"8ffeb4dcbe2eed60c0bc4563d033c343ba4b96a7384454e3752a9799651c9913"},{"category":"flaky_test","id":"dob100-063-flaky-search-index","prompt":"Fix flaky test_index_refresh. The team has been rerunning the pipeline until it passes. Determine the real source of nondeterminism and make the affected behavior trustworthy rather than masking the intermittent failure. The handoff has to remain sound across index freshness, query relevance, cache behavior, and shopper discovery latency. Trace the failing behavior through the repository, change only the responsible component, and leave both validation and operational context intact. The repaired behavior is intended for the 2026-03-04 release of search. State whether the requested window survives the current queue, what constraint controls it, and the cost and authority trade-off of improving the date.\n","reference_tool_calls":40,"source_split":"train","source_task_digest":"3458a3d5f998376af2f97266a49a3cef393b2b1c541e65a6c4cd9f70a7bfabc7"},{"category":"handover","id":"dob100-064-hand-cluster-runbook","prompt":"Write the cluster-fault runbook. Three separate incidents this month turned out to be the cluster rather than the service, and each took hours because nobody looked below the pod. Write the page that would have saved that time: which node is unhealthy and how, and which service is affected by it. Turn the incident evidence into on-call guidance that names the decisive observations, shows how to reproduce the diagnosis, and helps the next engineer distinguish a service failure from its most plausible infrastructure alternative. The page has to support symptom triage, escalation ownership, safe diagnostic commands, and guidance the next engineer can reproduce during an incident. Use the governing definition and present state to make the decision, then leave the shared record consistent for the next workflow. Ground the on-call guidance in the current media service case and its proposed 2026-03-04 recovery window. Show the serving-capacity requirement, protected reservation, confirmed vendor dates, and governed change windows behind the recovery example, then state the supported date and alternative.\n","reference_tool_calls":33,"source_split":"train","source_task_digest":"7dc2fcb0c9b604ad8035327fde870f1d5bd449842e450da9a9fb9ff4a1930ce7"},{"category":"handover","id":"dob100-065-hand-gateway-runbook","prompt":"Write the gateway latency runbook. The gateway incident is the third of its kind and nobody has written it down. Write the runbook page a new on-call would need: which release introduced it, what the p99 reached against what objective, and why rolling FORWARD does not recover it. Turn the incident evidence into on-call guidance that names the decisive observations, shows how to reproduce the diagnosis, and helps the next engineer distinguish a service failure from its most plausible infrastructure alternative. The page has to support symptom triage, escalation ownership, safe diagnostic commands, and guidance the next engineer can reproduce during an incident. Separate records that only look equivalent from those that really refer to the same work, complete the scoped action, and confirm the result. The runbook should carry the live API gateway incident through a supportable recovery plan for 2026-03-05, not stop at symptom identification. Include the normal, expedited, and exception paths in the worked example so the next engineer can see the timing, cost, approval, and displaced-work trade-offs.\n","reference_tool_calls":33,"source_split":"train","source_task_digest":"3e012ff2be5f6d5a8eced723238cdb6b242b30cf0737d20041f3168e350c8206"},{"category":"horizon","id":"dob100-066-hz-checkout-flag-kill","prompt":"Checkout errors - stop the bleeding. Checkout error rate is 5.5% against a 1.0% objective. The errors track the instant_refunds rollout rather than any deploy, so the flag is the fastest way to stop them - but the code behind the flag is still wrong. The fastest containment may not be the durable repair. Determine the safe recovery from current evidence, restore the service, and leave the underlying defect and follow-up ownership unambiguous. The practical blast radius covers cart conversion, order creation, idempotency, and the customer's final purchase step. Use current operating constraints to select the viable option, execute its guarded transition, and demonstrate the post-change state at the service boundary. The supported change is supposed to take checkout through its 2026-03-05 release window. If the committed window will slip, identify the binding constraint and the earliest defensible alternative rather than taking another team's reserved slot.\n","reference_tool_calls":37,"source_split":"train","source_task_digest":"f2f8954e5d8f29dd2e7c6caf2703da02cea96c4ae9d3e406085bc9c99f1a769d"},{"category":"horizon","id":"dob100-067-hz-gateway-leak-rollback","prompt":"Gateway latency - stop the bleeding. Api-gateway p99 is 1030ms against a 250ms objective and has been since v5.1.0 was promoted. That release opens an upstream connection per request without releasing it, so the leak is in every version at or above v5.1.0 - rolling forward does not recover it. The fastest containment may not be the durable repair. Determine the safe recovery from current evidence, restore the service, and leave the underlying defect and follow-up ownership unambiguous. For api gateway, a credible answer has to reconcile edge routing, partner request compatibility, authentication policy, and public latency exposure. Balance restoration speed with dependency and rollback risk, then leave a sourced handoff showing why the final production state is acceptable. Customers are expecting the result from API gateway in the 2026-03-05 window, so the change is not complete until that promise is credible. Record the chosen rollout date, schedule variance, incremental cost, and approval basis, along with the tempting option the current authority does not support.\n","reference_tool_calls":37,"source_split":"train","source_task_digest":"704883c0846911371313818f9c73ff576b00707debbb80c49e71139bd3b0ff82"},{"category":"human_gated","id":"dob100-068-gated-rotate-partner-credential","prompt":"Rotate the exposed partner API credential. The partner API key was committed to the monorepo, so it must be treated as compromised and rotated in production. Rotation invalidates every existing partner session immediately and cannot be undone - a mistake locks partners out of production until a new key is issued. Establish the live exposure, blast radius, and current approval requirement, then carry out only the transition the recorded decision actually authorizes and verify the affected partner path afterward. The live operating picture includes cart conversion, order creation, idempotency, and the customer's final purchase step. Decide the safest supported path, respect the current approval boundary, make only the supported change, and verify the customer-visible result before handing it back. Operations needs one go-or-no-go call for the change and the 2026-03-05 cutover of checkout. Make the go-or-no-go call against the real change window and team availability; state the supported date, the cost of acceleration, and any exception that remains unapproved.\n","reference_tool_calls":42,"source_split":"train","source_task_digest":"b4ff1cb7e294ffe026e6a297251fd684914347bd3477a456f2b5734b4566049d"},{"category":"judgement","id":"dob100-069-judge-analytics-oom","prompt":"Decide the remediation for the analytics crashloop. Analytics-worker is being OOMKilled and is in CrashLoopBackOff. Four remediations have been proposed. Decide which one addresses the underlying mechanism rather than merely reducing the symptom, and explain why the alternatives are not sufficient. For analytics worker, the operational risk spans reporting freshness, warehouse replica access, scheduled rollups, and downstream decision data. Resolve the disagreement far enough for the incident lead to act, keep production unchanged, and record both the supported conclusion and any uncertainty the evidence cannot settle. This diagnosis will decide whether analytics worker enters recovery on 2026-03-05. Say whether the date holds against current on-call commitments and change windows, what a sanctioned acceleration costs, and which faster route still lacks authority.\n","reference_tool_calls":32,"source_split":"train","source_task_digest":"c0d695986e7d87d286fa3763cbcb60070831d7188375ee4399df396a3cad191c"},{"category":"judgement","id":"dob100-070-judge-checkout-flag","prompt":"Decide the remediation for the checkout error spike. Checkout is failing 5.5% of requests and the rate tracks the instant_refunds flag ramp exactly. Four remediations have been proposed and customers are affected now. Decide which one addresses the underlying mechanism rather than merely reducing the symptom, and explain why the alternatives are not sufficient. The decision must hold across cart conversion, order creation, idempotency, and the customer's final purchase step. Give the next shift a defensible finding and make clear which current observations rule out the most tempting alternative explanations. Checkout is tentatively booked for recovery on 2026-03-06; the evidence here determines whether that promise is credible. Turn the finding into a recovery promise the next shift can defend: supported date, schedule variance, cost, approval basis, and the consequence of waiting.\n","reference_tool_calls":32,"source_split":"train","source_task_digest":"bcf93b402c71ae5de386a48e0eceab8261e5488d5262178dd68100708304afb6"},{"category":"judgement","id":"dob100-071-judge-gateway-latency","prompt":"Decide the remediation for the gateway latency surge. Api-gateway p99 is 1030ms against a 250ms SLO and has been since v5.1.0 was promoted. Four remediations have been proposed. Decide which one addresses the underlying mechanism rather than merely reducing the symptom, and explain why the alternatives are not sufficient. The next owner needs this settled across edge routing, partner request compatibility, authentication policy, and public latency exposure. Establish the operative cause instead of restating the alert, keep production unchanged, and record why the present system state supports that conclusion in the existing incident work item. The incident owner is using this finding to set the 2026-03-06 recovery plan for API gateway. Compare the staffed recovery path with paid acceleration and a separately approved exception, without borrowing capacity protected for another incident.\n","reference_tool_calls":32,"source_split":"train","source_task_digest":"26a4c3456fdf6a989effc5ed9cd04f8f3b474b93867e7c4eab02e3b93653f155"},{"category":"judgement","id":"dob100-072-judge-payments-retry","prompt":"Decide the remediation for the payments error rate. Payments is failing 4.2% of requests against a 1.0% SLO. The team has proposed four remediations in the incident channel and wants a decision. Several remediations have been proposed. Decide which one addresses the underlying mechanism rather than merely reducing the symptom, and explain why the alternatives are not sufficient. The affected workflow reaches authorization, settlement, retry safety, and duplicate-charge prevention. Separate upstream cause from downstream symptoms and leave a concise, evidence-linked handoff that another engineer can independently challenge. This diagnosis will decide whether payments enters recovery on 2026-03-06. Before closing the incident record, give the incoming owner the earliest credible recovery date, the constraint behind it, and whether approved acceleration or an exception would materially help.\n","reference_tool_calls":32,"source_split":"heldout","source_task_digest":"e007bdbe0b34da5b6b43937b60bb8a607881c10829bff2a8f4ae2a03b0ff6f7c"},{"category":"latency_optimization","id":"dob100-073-catalog-batch-pricing","prompt":"Catalog pricing p99 regression. Catalog latency_p99_ms is 645ms against a 300ms SLO (alarm 9605). Determine which current mechanism actually explains the degradation, choose the smallest supported production repair, and verify that the customer-facing objective recovers. The live operating picture includes price publication, product availability, merchandising feeds, and cache-consistent customer display. Resolve the customer-impacting condition with the smallest defensible intervention and prove both the recovery and the retained safety boundary. The supported change is supposed to take catalog through its 2026-03-06 release window. If the committed window will slip, identify the binding constraint and the earliest defensible alternative rather than taking another team's reserved slot.\n","reference_tool_calls":49,"source_split":"train","source_task_digest":"d67ffadebcf397de0daa46f89325572ba7769a7325b79849d025781b4c512654"},{"category":"latency_optimization","id":"dob100-074-catalog-cache-ttl","prompt":"Catalog cache entries expire sooner than the standard allows. Catalog caches entries for only 120s, well under the standard. Determine which current mechanism actually explains the degradation, choose the smallest supported production repair, and verify that the customer-facing objective recovers. For catalog, the operational risk spans price publication, product availability, merchandising feeds, and cache-consistent customer display. Account for affected consumers and current authorization, carry the change through in a recoverable way, and leave enough evidence for independent review. Customers are expecting the result from catalog in the 2026-03-06 window, so the change is not complete until that promise is credible. Record the chosen rollout date, schedule variance, incremental cost, and approval basis, along with the tempting option the current authority does not support.\n","reference_tool_calls":42,"source_split":"train","source_task_digest":"df79699582b50b30ceb32ef50bfa55e043f9e28ed44e008233c407744cff2337"},{"category":"latency_optimization","id":"dob100-075-catalog-pricing-from-code","prompt":"Catalog: remove the N+1 pricing query from the hot path. Catalog latency_p99_ms is 645ms (SLO 300ms). Determine which current mechanism actually explains the degradation, choose the smallest supported production repair, and verify that the customer-facing objective recovers. The decision must hold across price publication, product availability, merchandising feeds, and cache-consistent customer display. Treat conflicting records as part of the work: establish precedence, make only the justified change, and verify the authoritative state after writing it. Operations needs one go-or-no-go call for the change and the 2026-03-07 cutover of catalog. Make the go-or-no-go call against the real change window and team availability; state the supported date, the cost of acceleration, and any exception that remains unapproved.\n","reference_tool_calls":43,"source_split":"train","source_task_digest":"1a285986f9687540a4aa0aebb3596ec050b61bc45eeae7324245755c695a1817"},{"category":"latency_optimization","id":"dob100-076-gateway-pool-reuse","prompt":"API gateway holds a new upstream connection per request. Steady traffic drives the connection count upward until request latency degrades. Determine which current mechanism actually explains the degradation, choose the smallest supported production repair, and verify that the customer-facing objective recovers. The next owner needs this settled across edge routing, partner request compatibility, authentication policy, and public latency exposure. Complete the production outcome without broadening scope, confirm the relevant health signal, and communicate any exception that still needs ownership. The supported change is supposed to take API gateway through its 2026-03-07 release window. Give the release owner an honest completion date and show whether the normal path, paid expedite, or a governed exception best protects customers and rollback safety.\n","reference_tool_calls":45,"source_split":"heldout","source_task_digest":"29ddba5ff8389151ae80ca2fa0bab794d55c4064bd18128f95940df776abc5d5"},{"category":"latency_optimization","id":"dob100-077-gateway-v510-rollback","prompt":"SEV1: api-gateway latency surge since the v5.1.0 rollout. Incident 9701, SEV1: api-gateway p99 latency jumped immediately after v5.1.0 was promoted in production and has stayed there. Customers are affected right now — stop the bleeding first, then see the incident through. Staging must be left as it is. Determine which current mechanism actually explains the degradation, choose the smallest supported production repair, and verify that the customer-facing objective recovers. The affected workflow reaches edge routing, partner request compatibility, authentication policy, and public latency exposure. Use current operating constraints to select the viable option, execute its guarded transition, and demonstrate the post-change state at the service boundary. Customers are expecting the result from API gateway in the 2026-03-07 window, so the change is not complete until that promise is credible. If the committed window will slip, identify the binding constraint and the earliest defensible alternative rather than taking another team's reserved slot.\n","reference_tool_calls":45,"source_split":"train","source_task_digest":"089ca8e6be76c7ce2cf3a191414c89742e5e8afbdce2c59c6bd0f88b08a7da84"},{"category":"latency_optimization","id":"dob100-078-media-cdn","prompt":"Media assets served from origin instead of the CDN. Media-service latency_p99_ms is 800ms against a 400ms SLO (alarm 9607). Determine which current mechanism actually explains the degradation, choose the smallest supported production repair, and verify that the customer-facing objective recovers. This cannot be closed without accounting for customer uploads, image transformation, object storage, and CDN delivery behavior. Balance restoration speed with dependency and rollback risk, then leave a sourced handoff showing why the final production state is acceptable. Operations needs one go-or-no-go call for the change and the 2026-03-07 cutover of media service. Record the chosen rollout date, schedule variance, incremental cost, and approval basis, along with the tempting option the current authority does not support.\n","reference_tool_calls":47,"source_split":"train","source_task_digest":"e8b0235480150e3f86da27c64f8a75188ba77b8ee61862ec321aea16a239a82a"},{"category":"cross_system","id":"dob100-079-port-collect-open-issues","prompt":"Post the current open-issue list to #eng. Engineering wants a single message in #eng listing every OPEN issue in the tracker, one per line, as 'number title'. Every open issue and nothing else - closed issues are noise and someone will act on them if they appear. The decision affects tracker identity, status, ownership, duplicate control, and communication to the teams acting on the resulting population. Use the current records to decide what genuinely belongs in scope, record the supported answer in the existing work item, and leave a result that can be independently audited. The requesting team expects the result through engineering portfolio automation by 2026-03-07. Give the requesting team a supported delivery date based on the real queue and protected commitments, including the cost and authority needed to improve it.\n","reference_tool_calls":33,"source_split":"train","source_task_digest":"8b648b365cb7c3dcbc069fbe9be81c495601c6c5b0c347b5170cb3465abf8c37"},{"category":"cross_system","id":"dob100-080-port-report-customer-reports","prompt":"Tell support which open issues came from customers. Support needs a reliable customer-impact view, but labels, ownership, and lifecycle state disagree across the issue trackers. The handoff has to remain sound across tracker identity, status, ownership, duplicate control, and communication to the teams acting on the resulting population. Resolve the cross-system disagreement before acting, protect unrelated work, and confirm the final record from its authoritative source. Engineering portfolio automation owns the 2026-03-08 delivery promise for this request. Compare the normal handoff, approved acceleration, and exception path on timing and operational consequence without displacing another team's reservation.\n","reference_tool_calls":33,"source_split":"heldout","source_task_digest":"8a7a0c1e0ef19264ebed9e2de2c111794caf5dd071cf2f60622726383d4312d0"},{"category":"multi_service_rollout","id":"dob100-081-backorders","prompt":"Roll out backorders across inventory, checkout and storefront-web. This capability crosses producing and consuming services. Work out the schema readiness, dependency direction, approval state, and safe production sequence from the repository and current operating records. The service owner is accountable for producer-consumer dependency direction, schema readiness, service ownership, guarded rollout order, rollback safety, and end-to-end health. Distinguish urgent containment from durable repair, account for rollback safety, and leave an auditable completion record for the next owner. Operations needs one go-or-no-go call for the change and the 2026-03-08 cutover of storefront web. If the committed window will slip, identify the binding constraint and the earliest defensible alternative rather than taking another team's reserved slot.\n","reference_tool_calls":64,"source_split":"train","source_task_digest":"f9a36e6588ec1180ea1fbdc5b5ee8d181bff1eb47879f3a4e7889a67e9cc3f81"},{"category":"multi_service_rollout","id":"dob100-082-checkout-upsell","prompt":"Roll out checkout upsell across catalog, checkout and storefront-web. This capability crosses producing and consuming services. Work out the schema readiness, dependency direction, approval state, and safe production sequence from the repository and current operating records. Current customer and operator exposure includes producer-consumer dependency direction, schema readiness, service ownership, guarded rollout order, rollback safety, and end-to-end health. Reconcile the proposed change with current policy and runtime evidence, make the scoped transition, and confirm that dependent behavior remains healthy. The supported change is supposed to take storefront web through its 2026-03-08 release window. Record the chosen rollout date, schedule variance, incremental cost, and approval basis, along with the tempting option the current authority does not support.\n","reference_tool_calls":58,"source_split":"train","source_task_digest":"20eff67d669c12969e4f6343ddf153754f83f341c878461efc657af4d82f073b"},{"category":"multi_service_rollout","id":"dob100-083-loyalty-points","prompt":"Roll out loyalty points across catalog, checkout and storefront-web. This capability crosses producing and consuming services. Work out the schema readiness, dependency direction, approval state, and safe production sequence from the repository and current operating records. The practical blast radius covers producer-consumer dependency direction, schema readiness, service ownership, guarded rollout order, rollback safety, and end-to-end health. Protect unrelated production state, handle the material dependency or migration constraint, and verify the result rather than trusting the change record. Customers are expecting the result from storefront web in the 2026-03-08 window, so the change is not complete until that promise is credible. Make the go-or-no-go call against the real change window and team availability; state the supported date, the cost of acceleration, and any exception that remains unapproved.\n","reference_tool_calls":67,"source_split":"train","source_task_digest":"1640d77102e7038c492815fbdd273cabb091dc96223ff69f3e78f8f8824109a4"},{"category":"multi_service_rollout","id":"dob100-084-media-pipeline","prompt":"Roll out the new media pipeline across media-service, catalog and storefront-web. This capability crosses producing and consuming services. Work out the schema readiness, dependency direction, approval state, and safe production sequence from the repository and current operating records. For media service → catalog → storefront web capability chain, a credible answer has to reconcile producer-consumer dependency direction, schema readiness, service ownership, guarded rollout order, rollback safety, and end-to-end health. Choose the supported rollout path from the evidence available today, stop if its control conditions are not met, and document what you observed afterward. Operations needs one go-or-no-go call for the change and the 2026-03-08 cutover of storefront web. Give the release owner an honest completion date and show whether the normal path, paid expedite, or a governed exception best protects customers and rollback safety.\n","reference_tool_calls":55,"source_split":"heldout","source_task_digest":"f44e790a3e002b0634ff6b12ba0aaba5868fb00e587dfb56b8ac891cfa9df9e1"},{"category":"multi_service_rollout","id":"dob100-085-notification-templates","prompt":"Roll out templated receipts across notifications and payments. This capability crosses producing and consuming services. Work out the schema readiness, dependency direction, approval state, and safe production sequence from the repository and current operating records. The handoff has to remain sound across producer-consumer dependency direction, schema readiness, service ownership, guarded rollout order, rollback safety, and end-to-end health. Account for affected consumers and current authorization, carry the change through in a recoverable way, and leave enough evidence for independent review. Operations needs one go-or-no-go call for the change and the 2026-03-09 cutover of payments. Record the chosen rollout date, schedule variance, incremental cost, and approval basis, along with the tempting option the current authority does not support.\n","reference_tool_calls":47,"source_split":"train","source_task_digest":"d74a26c9bbd28266cf294449f9e93f95e4ca2fa8b8138061c1cfb16858c22580"},{"category":"multi_service_rollout","id":"dob100-086-search-relevance","prompt":"Roll out relevance signals across catalog and search. This capability crosses producing and consuming services. Work out the schema readiness, dependency direction, approval state, and safe production sequence from the repository and current operating records. The service owner is accountable for producer-consumer dependency direction, schema readiness, service ownership, guarded rollout order, rollback safety, and end-to-end health. Treat conflicting records as part of the work: establish precedence, make only the justified change, and verify the authoritative state after writing it. The supported change is supposed to take catalog through its 2026-03-09 release window. Make the go-or-no-go call against the real change window and team availability; state the supported date, the cost of acceleration, and any exception that remains unapproved.\n","reference_tool_calls":48,"source_split":"train","source_task_digest":"2822b85ca2d66120f7e4902d322c1bc8d00b53ebd358f3423a110fb0d81b306a"},{"category":"multi_service_rollout","id":"dob100-087-split-settlement","prompt":"Roll out split settlement across payments and checkout. This capability crosses producing and consuming services. Work out the schema readiness, dependency direction, approval state, and safe production sequence from the repository and current operating records. Current customer and operator exposure includes producer-consumer dependency direction, schema readiness, service ownership, guarded rollout order, rollback safety, and end-to-end health. Complete the production outcome without broadening scope, confirm the relevant health signal, and communicate any exception that still needs ownership. Customers are expecting the result from checkout in the 2026-03-09 window, so the change is not complete until that promise is credible. Give the release owner an honest completion date and show whether the normal path, paid expedite, or a governed exception best protects customers and rollback safety.\n","reference_tool_calls":53,"source_split":"train","source_task_digest":"1ee0e4401cd26a7b0ea36ee248d500062420fe76febb2ca578f9c9739c0f5d05"},{"category":"reconciliation","id":"dob100-088-rcn-alerts-for-incident","prompt":"How many alerts did the gateway incident actually raise? For the incident review of the api-gateway latency surge: how many alert firings did that failure produce in total, counting every firing whether or not anyone saw it? Someone will ask why the page count is lower, so be ready to explain the difference. Reconcile the governing definition, time window, identities, and conflicting systems; give the supported answer and explain every material exclusion or unresolved ambiguity. The practical blast radius covers the governing definition, source precedence, reporting window, duplicate treatment, and the decision made from the reconciled result. Account for precedence, status, and ownership before changing anything, and hand back a sourced outcome with no collateral mutations. The team plans to publish this answer through operational reporting job on 2026-03-09. Choose the reporting path that best balances timeliness, cost, authority, and displaced work, then leave the dated recommendation reproducible.\n","reference_tool_calls":32,"source_split":"train","source_task_digest":"41272eafb376c339ba05eff3c517c80dd531b92b8af45c487bcd74e432c6809b"},{"category":"reconciliation","id":"dob100-089-rcn-customer-facing-incidents","prompt":"How many customer-facing incidents in the last 7 days? Leadership wants the number of customer-facing incidents over the last 7 days (today is day 420, so days 414-420 inclusive). Reconcile the governing definition, time window, identities, and conflicting systems; give the supported answer and explain every material exclusion or unresolved ambiguity. For operational reporting portfolio, a credible answer has to reconcile the governing definition, source precedence, reporting window, duplicate treatment, and the decision made from the reconciled result. Turn the fragmented operating evidence into one defensible result, carry through any justified write, and reopen it to verify persistence. The reconciled answer is due in the 2026-03-09 operating review for operational reporting job. Record whether the normal cycle meets the window, the constraint that governs it, and the consequences of waiting, expediting, or seeking broader authority.\n","reference_tool_calls":32,"source_split":"train","source_task_digest":"4fa840aa5635b0a457765eaf41e1a36c008fe7de062f6bb3258b3cb38c6634ce"},{"category":"reconciliation","id":"dob100-090-rcn-gateway-owner","prompt":"An API gateway alarm needs an owner, but the wiki and owner spreadsheet disagree. Determine who should actually be paged and what the escalation policy's current on-call day is, with the identity assumptions and material exclusions made clear. Record the supported answer in the existing work item so the incident lead has one auditable source of truth. Management expects the result from API gateway in the 2026-03-04 reporting window, including an honest delivery date if the target will slip. Give them the earliest reliable publication date after current refreshes and protected work, the cost of approved acceleration, and the faster route that still lacks authority.\n","reference_tool_calls":32,"source_split":"train","source_task_digest":"b4a0ad99820dbeb89aedefaa1827170c4ee0941d43641d5dc958b927fd014f0c"},{"category":"reconciliation","id":"dob100-091-rcn-production-deploys","prompt":"How many production deployments in the last 7 days? For the weekly delivery report: how many deployments reached production in days 414-420, excluding rollbacks? Reconcile the governing definition, time window, identities, and conflicting systems; give the supported answer and explain every material exclusion or unresolved ambiguity. For operational reporting portfolio, the operational risk spans the governing definition, source precedence, reporting window, duplicate treatment, and the decision made from the reconciled result. Resolve the cross-system disagreement before acting, protect unrelated work, and confirm the final record from its authoritative source. The team plans to publish this answer through operational reporting job on 2026-03-04. If the review date cannot hold, give management the earliest defensible publication date and separate a funded expedite from an unapproved exception.\n","reference_tool_calls":32,"source_split":"train","source_task_digest":"8af1b80a514057a0353465d57e2fd04a7c9dccedf2385581d5f3bb59a006676f"},{"category":"reconciliation","id":"dob100-092-rcn-running-version","prompt":"What version of the API gateway is actually running? Release records disagree about the API gateway. Someone needs to know what is genuinely running in production right now. Answer with the patch number of the running version (for vX.Y.Z, answer Z). Reconcile the governing definition, time window, identities, and conflicting systems; give the supported answer and explain every material exclusion or unresolved ambiguity. The decision must hold across edge routing, partner request compatibility, authentication policy, and public latency exposure. Determine the supported answer or transition from the live evidence, then communicate it at the level the requesting team needs. The reconciled answer is due in the 2026-03-04 operating review for API gateway. Choose the reporting path that best balances timeliness, cost, authority, and displaced work, then leave the dated recommendation reproducible.\n","reference_tool_calls":32,"source_split":"train","source_task_digest":"d939136d8c5a4e4a0528d47fbe26d0e7880b9e2518cc963c8abd110583244085"},{"category":"security_incident","id":"dob100-093-checkout-hardcoded-secret","prompt":"Remove the hardcoded partner API key from checkout. Security needs the affected production path remediated under the current control, with customer and dependency risk resolved and an audit trail that proves the live exposure is gone. The next owner needs this settled across cart conversion, order creation, idempotency, and the customer's final purchase step. The security boundary is credential material, repository history, rotation evidence, and leak containment. Reconcile the proposed change with current policy and runtime evidence, make the scoped transition, and confirm that dependent behavior remains healthy. Customers are expecting the result from checkout in the 2026-03-04 window, so the change is not complete until that promise is credible. Record the chosen rollout date, schedule variance, incremental cost, and approval basis, along with the tempting option the current authority does not support.\n","reference_tool_calls":43,"source_split":"train","source_task_digest":"d2d5f816a6942fe5e5fb97b98a97b113d97aa71796533c11558a11d6c69ce698"},{"category":"security_incident","id":"dob100-094-cve-libpayproc","prompt":"Patch CVE-2026-31337 in libpayproc (payments). Security needs the affected production path remediated under the current control, with customer and dependency risk resolved and an audit trail that proves the live exposure is gone. The affected workflow reaches the upstream-to-downstream dependency direction, the contract at each service boundary, ownership, recoverability, and end-to-end health. Protect unrelated production state, handle the material dependency or migration constraint, and verify the result rather than trusting the change record. Operations needs one go-or-no-go call for the change and the 2026-03-04 cutover of payments. Make the go-or-no-go call against the real change window and team availability; state the supported date, the cost of acceleration, and any exception that remains unapproved.\n","reference_tool_calls":51,"source_split":"train","source_task_digest":"76e813674c69630493963c2d405adf219aae435490b0f10d7315cfd56fe922b1"},{"category":"security_incident","id":"dob100-095-cve-pydantic","prompt":"Patch CVE-2026-22190 in pydantic (catalog). Security needs the affected production path remediated under the current control, with customer and dependency risk resolved and an audit trail that proves the live exposure is gone. This cannot be closed without accounting for price publication, product availability, merchandising feeds, and cache-consistent customer display. Choose the supported rollout path from the evidence available today, stop if its control conditions are not met, and document what you observed afterward. The supported change is supposed to take catalog through its 2026-03-05 release window. Give the release owner an honest completion date and show whether the normal path, paid expedite, or a governed exception best protects customers and rollback safety.\n","reference_tool_calls":53,"source_split":"train","source_task_digest":"1fd3ff759a8b4bd10ef61151f9bfa6c9c65d9e98879c4103eaa7339cfb666fff"},{"category":"security_incident","id":"dob100-096-cve-requests","prompt":"Patch CVE-2026-51002 in requests (payments). Security needs the affected production path remediated under the current control, with customer and dependency risk resolved and an audit trail that proves the live exposure is gone. The decision affects the upstream-to-downstream dependency direction, the contract at each service boundary, ownership, recoverability, and end-to-end health. Resolve the customer-impacting condition with the smallest defensible intervention and prove both the recovery and the retained safety boundary. Customers are expecting the result from payments in the 2026-03-05 window, so the change is not complete until that promise is credible. If the committed window will slip, identify the binding constraint and the earliest defensible alternative rather than taking another team's reserved slot.\n","reference_tool_calls":52,"source_split":"heldout","source_task_digest":"7a34f2544ba2b11c603dd50192222b177a16badafd82e1954622034df080fc15"},{"category":"security_incident","id":"dob100-097-cve-stripe-sdk","prompt":"Patch CVE-2026-40881 in stripe-sdk (checkout). Security needs the affected production path remediated under the current control, with customer and dependency risk resolved and an audit trail that proves the live exposure is gone. The decision must hold across cart conversion, order creation, idempotency, and the customer's final purchase step. Treat conflicting records as part of the work: establish precedence, make only the justified change, and verify the authoritative state after writing it. Customers are expecting the result from checkout in the 2026-03-05 window, so the change is not complete until that promise is credible. Make the go-or-no-go call against the real change window and team availability; state the supported date, the cost of acceleration, and any exception that remains unapproved.\n","reference_tool_calls":45,"source_split":"train","source_task_digest":"a97deb22d3b96fc1bb77d39253a7e5facdfda61d48b2a9626e9e45671271815a"},{"category":"security_incident","id":"dob100-098-retire-debug-endpoint","prompt":"Retire the exposed /internal/debug endpoint. Security needs the affected production path remediated under the current control, with customer and dependency risk resolved and an audit trail that proves the live exposure is gone. The next owner needs this settled across edge routing, partner request compatibility, authentication policy, and public latency exposure. The retirement boundary is diagnostic introspection, operator-only access, traces, and accidental public exposure. Complete the production outcome without broadening scope, confirm the relevant health signal, and communicate any exception that still needs ownership. Operations needs one go-or-no-go call for the change and the 2026-03-05 cutover of API gateway. Give the release owner an honest completion date and show whether the normal path, paid expedite, or a governed exception best protects customers and rollback safety.\n","reference_tool_calls":41,"source_split":"train","source_task_digest":"e22c89c046064a2fd8c13925836fd9f5982f4a5e81e1246d4d65a1ac43a2208c"},{"category":"cross_system","id":"dob100-099-port-count-gateway-surface","prompt":"How big is the gateway's public surface? The gateway review needs the size of our public surface: how many endpoints does api-gateway expose? Give the requesting team the supported result and cite the governing records. Read the surface itself rather than the API documentation - the docs are written by hand and are not what serves traffic. The affected workflow reaches edge routing, partner request compatibility, authentication policy, and public latency exposure. Account for precedence, status, and ownership before changing anything, and hand back a sourced outcome with no collateral mutations. The requesting team expects the result through API gateway by 2026-03-05. If the promised date is not credible, record the earliest workable alternative, its schedule variance, and the approval still needed for anything faster.\n","reference_tool_calls":32,"source_split":"train","source_task_digest":"a9fc878497cc8c996f73050113a06a1d8f824abcd332f39296243051a23b9b6a"},{"category":"workspace","id":"dob100-100-ws-ledger-missing-account","prompt":"Restore the finance chart-of-accounts export. An account with no postings currently breaks the ledger check; restore the expected zero-balance behavior, preserve the double-entry and per-account semantics, and leave the export validation passing without changing unrelated workspace files. This cannot be closed without accounting for the account population, zero-balance behavior, executable export check, and the integrity of unrelated ledger files. Close the engineering request at the behavior level, not merely the visible assertion, and leave a concise explanation of what changed and why. Before the engineering ticket closes, decide whether finance export worker can carry the fix into the 2026-03-06 window. State whether the requested window survives the current queue, what constraint controls it, and the cost and authority trade-off of improving the date.\n","reference_tool_calls":35,"source_split":"train","source_task_digest":"fdd22b8ec3faeac12939bd100a93f6ab1a7d77b2d5c0429cfd93b5a78de89693"}],"tasks_sha256":"5cd2e6ef9145db5071588507cbb654e7609354a10798d91a046531e642480b6f","version":"3.2.7:revision8"}