182 lines
46 KiB
JSON
182 lines
46 KiB
JSON
{"prompt": "- new AppIntent for \"start a focus session\"\n- needs to work from spotlight and shortcuts\n- parameter for duration, default 25m\n- return a snippet view showing the timer\n- and it should deep link into the app if you tap it\n\nswiftui, ios 18 target", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "the menu bar extra should show a live count and a popover with the last 5 items, plus a preference for launch at login", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "sqlite is fine for now but at some point we need real concurrency. write up when we'd have to move, to what, and what our schema does that would make it painful", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "shell completions for bash zsh and fish, generated from the command tree not hand written", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "so the tricky part is that `deploy --watch` needs to keep streaming logs while also handling ctrl-c cleanly and not leaving the remote build orphaned... anyway implement it", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "half our funcs return (T, error) and half return a custom Result. unify on the idiomatic go one, no behavior change", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "rename --dir to --path but keep --dir working with a deprecation warning", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.3, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "index bloat on the audit table, 40gb of index on 12gb of data. what do we do about it and in what order", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "why does EXPLAIN say rows=1 when there are clearly 40k matching", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "walk me through what our vacuum settings actually mean for the events table, autovacuum_vacuum_scale_factor is 0.01 and i don't remember why", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "man page for the cli. every flag, exit codes, three examples at the bottom, standard troff sections", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.4, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "the Info.plist is missing NSMicrophoneUsageDescription and the app store rejected us", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.1, "slice": "core", "lang": "en"}
|
|
{"prompt": "our Core Data stack, the CloudKit sync, and the view models are all tangled through NSManagedObject subclasses in the views. put a domain model layer in between. the ui must behave exactly the same, including the animations on insert", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "keyboard shortcuts in the mac app: cmd+shift+n new window, cmd+opt+arrows to move between panes, and they need to show in the menu", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "table has 80M rows and no partitioning. plan the move to monthly partitions with no downtime, including how we handle the existing indexes and the fk from line_items", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.9, "slice": "core", "lang": "en"}
|
|
{"prompt": "postgres advisory locks vs a lock table for our job claiming — which do we already use, and where. just tell me what's there", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "the app hangs when you open the third window, spindump attached mentally: everything's waiting on a main-thread dispatch that never returns. this started when i added the shared ModelContainer", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "wrong results from the group by when a tenant has no rows in the date range — we get a row with count 0 for the WRONG tenant sometimes. i've read the query 20 times", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "put together the docs for our sql style guide — cte naming, when to use lateral joins, why we ban select *", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.4, "slice": "core", "lang": "en"}
|
|
{"prompt": "extract the argument parsing out of main.go, it's 400 lines and impossible to test. same flags, same help text byte for byte", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "sheet presentation detents, medium and large, and the drag indicator should be hidden when it's at large", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.4, "slice": "core", "lang": "en"}
|
|
{"prompt": "the --verbose flag prints to stdout, should be stderr", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.2, "slice": "core", "lang": "en"}
|
|
{"prompt": "review the schema in db/schema.sql and tell me where we're going to regret our choices in a year", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "write the tutorial for building a first plugin, end to end, someone should have a working thing in 20 minutes", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "wrap it up", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.3, "slice": "vague-eval", "lang": "en"}
|
|
{"prompt": "goal: cut cold start on the lambda from 2.4s to under 500ms\nconstraints: python, we need pandas (yes really), and we can't move to a container image this quarter\nwant: a plan with the options ranked by effort vs payoff, and tell me which ones are dead ends", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "presigned upload flow for user avatars — validate content type server side after the upload, reject anything that isn't actually an image, and generate the 3 sizes", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "notification center in the web app: unread dot, mark all read, grouped by day, infinite scroll, and it should update live over the existing socket", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "the ttl on the session cache should be 24h not 7d", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.1, "slice": "core", "lang": "en"}
|
|
{"prompt": "we call it `slug` in the db, `handle` in the api, and `username` in the ui. pick `handle`, make it consistent through all three layers", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "intermittent 'database is locked' from sqlite in the desktop app. WAL is on, one writer, i thought that was supposed to be enough", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "is there any way a user can end up with two active subscriptions? read the checkout + webhook paths and tell me", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "our api guide says rate limits are per-key. they're per-ip in the code. update the guide and the error message copy to match reality", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.3, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "wondering if we should just... ok, question: how does the retry queue know not to reprocess an event that a different worker already handled", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "unread badge sync between the widget and the app is off by one after you read something in the widget", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "implement soft rate limiting on the graphql endpoint by query complexity, not request count. reject over 1000 points with a useful message", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "empty states for all six list screens, consistent illustration + copy pattern, and each one needs a different primary action", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "bump the sdk to 2.14.0 and see if anything breaks", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.3, "slice": "core", "lang": "en"}
|
|
{"prompt": "sort out the layering in internal/ — right now the store imports the handler package for a type, which is backwards. move the type, nothing else changes", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "roadmap for the offline-first rewrite of the notes app. three months, two devs, and we can't break existing users' data", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.9, "slice": "core", "lang": "en"}
|
|
{"prompt": "clicking a search result opens the wrong document about 5% of the time, and i can't reproduce it locally at all", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "what's actually different between our staging and prod postgres configs and does any of it matter for query plans", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "notes for the app review submission — what changed, what to test, and the demo account credentials placeholder", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.2, "slice": "core", "lang": "en"}
|
|
{"prompt": "here's the failing output, i think it's a one line change somewhere in the formatter:\n\n$ acmectl fmt --check config/\nconfig/prod.yaml\n--- expected\n+++ actual\n@@ -12,7 +12,7 @@\n replicas: 3\n resources:\n limits:\n- memory: \"2Gi\"\n+ memory: 2Gi\n requests:\n- memory: \"1Gi\"\n+ memory: 1Gi\n \n1 file would be reformatted, 22 files already formatted\nexit status 1\n\nour formatter is stripping the quotes off values that look like numbers-with-suffixes, which k8s then reads differently. it should leave quoted scalars alone", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.4, "slice": "pasted-context", "lang": "en"}
|
|
{"prompt": "figure out the plan for encrypting the notes at rest with a user passphrase, then implement the keychain part", "purpose": "planning", "secondary": "backendImpl", "mixed": true, "difficulty": 0.9, "slice": "mixed", "lang": "en"}
|
|
{"prompt": "tidy the SettingsView hierarchy and while you're in there fix the toggle that doesn't persist", "purpose": "refactor", "secondary": "quickFix", "mixed": true, "difficulty": 0.5, "slice": "mixed", "lang": "en"}
|
|
{"prompt": "one more pass on the importer", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "vague-eval", "lang": "en"}
|
|
{"prompt": "batch job to recompute the denormalized counters nightly, and a --dry-run that reports drift without writing", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "make the drag and drop between kanban columns work on touch, currently mouse only", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "el índice unique en (email) debería ser en (lower(email)), tenemos duplicados con mayúsculas", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.3, "slice": "boundary", "lang": "es"}
|
|
{"prompt": "explain to me, like i'm new here, how a request gets from the load balancer to a handler including the two middleware layers nobody documented", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "consolidate the four date helper files into one module. there's overlap and at least two of them handle timezones differently — pick the correct behavior and note where callers change", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "sequence the work for adding audit logs to every mutation. 60 mutations, and we want it enforced by types so a new one can't skip it", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "user says the app deleted their data. the sync log shows a tombstone we didn't send. i need to know how that's possible", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.9, "slice": "core", "lang": "en"}
|
|
{"prompt": "faq section for the docs based on the last 3 months of support tickets in support/exports/, group by theme", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "pagination on the admin users list, server side, 50 per page, keep the search query when you page", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.4, "slice": "core", "lang": "en"}
|
|
{"prompt": "the health check hits the db on every request, cache it for 5 seconds", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.3, "slice": "core", "lang": "en"}
|
|
{"prompt": "read the new indexing strategy doc in docs/ and tell me whether it holds up against our actual query patterns in queries/", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "implement two-phase deletes for accounts — 30 day grace, anonymize immediately, purge after, and everything queryable during the grace period must exclude them", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "get rid of the `_v2` suffix on the four services that have it, nothing is on v1 anymore", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.4, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "our postgres connections spike to the max every morning at 9:02 and recover by 9:06. pooler is pgbouncer in transaction mode. what is doing that", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "write the docs for the new `--profile` flag including how it interacts with the env vars and the config file, precedence order especially", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.4, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "liquid glass sidebar treatment on macOS 26, but keep the legacy material path for older versions", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "design how we'd let customers bring their own s3 bucket for exports. credentials, validation, what happens when they revoke access mid-job", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "eslint --fix left 3 things it can't fix, take a look", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.2, "slice": "core", "lang": "en"}
|
|
{"prompt": "does the transaction in `transferFunds` actually roll back the ledger entry if the notification send fails? trace it for me", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "here's the thing that's been driving me nuts:\n\n2026-07-28T14:22:09.114Z INFO [importer] batch 41 start rows=5000\n2026-07-28T14:22:09.882Z INFO [importer] batch 41 upserted=5000 skipped=0\n2026-07-28T14:22:10.004Z INFO [importer] batch 42 start rows=5000\n2026-07-28T14:22:41.229Z WARN [importer] batch 42 slow: 31.2s\n2026-07-28T14:22:41.230Z INFO [importer] batch 42 upserted=4998 skipped=2\n2026-07-28T14:22:41.401Z INFO [importer] batch 43 start rows=5000\n2026-07-28T14:22:42.109Z INFO [importer] batch 43 upserted=5000 skipped=0\n\nevery ~40th batch takes 30+ seconds and skips exactly 2 rows. the rest are sub-second. i've checked for hot keys, nothing obvious. skipped means the on-conflict did nothing", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "pasted-context", "lang": "en"}
|
|
{"prompt": "write the section of the readme about local development on apple silicon, the docker-compose needs platform overrides and nobody knows that", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.3, "slice": "core", "lang": "en"}
|
|
{"prompt": "pull-to-refresh on the timeline plus a small 'new posts' pill that appears when items arrive while you're scrolled down", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "webhook signature verification, hmac sha256, constant time compare, tolerate 5 min clock skew, reject replays with a nonce cache", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "hoist the feature flag checks scattered through the render tree into one place at the route level. same features on/off for the same users", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "still broken", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "vague-eval", "lang": "en"}
|
|
{"prompt": "we need a position on how to do multi-tenancy in the new service: schema per tenant, row level, or separate databases. write the analysis with our actual constraints — 4000 tenants, biggest one is 40% of the data", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.9, "slice": "core", "lang": "en"}
|
|
{"prompt": "grpc gateway in front of the internal services so the mobile team gets rest. codegen not hand written", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "the video player controls should auto hide after 3s of no interaction, fade not snap, and stay up while the scrubber is being dragged", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "gitignore the .DS_Store files that snuck in and remove them from the index", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.1, "slice": "core", "lang": "en"}
|
|
{"prompt": "the two implementations of exponential backoff (one in the http client, one in the queue consumer) should be one. the queue one is the correct one, it has jitter", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "requests hang for exactly 60s then succeed, roughly 1 in 50, only from our eu pods to the us database. dns? tcp? no idea", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.9, "slice": "core", "lang": "en"}
|
|
{"prompt": "assess whether the new auth library we're considering (openfga) actually fits our permission model. read models/permissions.go first", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "release notes 2026.8, and there's a behavior change in how we resolve relative paths that people will hit, lead with that", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.3, "slice": "core", "lang": "en"}
|
|
{"prompt": "wire up push notifications on android — fcm, topic per project, deep links into the right screen, and a quiet hours setting that's respected server side", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "tighten up the spacing in the compact table density, rows should be 32px not 40, and the checkbox column shouldn't shift", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.3, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "our github action uses ubuntu-20.04 which is retiring, move to 24.04", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.2, "slice": "core", "lang": "en"}
|
|
{"prompt": "restructure the test suite into unit/integration/e2e directories with separate make targets. same tests, same assertions, nothing skipped", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "audit our dependency tree for anything unmaintained or with a weird license, we're doing a compliance thing", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "guide for writing a custom exporter, aimed at people who've never touched our internals", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "memory on the ios app climbs 5mb per screen push and never comes back down. i suspect a retain cycle in the coordinator but instruments is showing me a wall of Malloc blocks and no obvious cycle", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "phased plan for introducing typescript to the 60k line js codebase. allowJs, then per-directory strictness, i think? tell me the right order and where the pain is", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "just do whatever's next on the list", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.4, "slice": "vague-eval", "lang": "en"}
|
|
{"prompt": "从 REST 迁移到 gRPC 的方案，先不写代码。我们有 12 个内部服务，客户端 SDK 是 TS 和 Kotlin", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "zh"}
|
|
{"prompt": "streaming json parser for the ndjson api responses so we don't hold 400mb in memory. rust, and it needs to handle a truncated final line", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "the settings sidebar in the electron app should collapse to icons under 900px wide, with tooltips", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.4, "slice": "core", "lang": "en"}
|
|
{"prompt": "we log at info in the hot loop, drop that one to debug", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.1, "slice": "core", "lang": "en"}
|
|
{"prompt": "the naming in the payments code is a mess — `txn`, `transaction`, `payment`, `charge` all mean the same thing in different files. standardize on `charge` since that's what the provider calls it", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "we get 'context canceled' errors on maybe 1% of writes and the data doesn't land, but the client sees a 200. that combination shouldn't be possible", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.9, "slice": "core", "lang": "en"}
|
|
{"prompt": "is there a reason the worker pool size is 7? just want to know if that's meaningful before i change it", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.3, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "prose for the pricing page's fine print about api limits, legal-adjacent but readable", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.3, "slice": "core", "lang": "en"}
|
|
{"prompt": "csv import wizard: upload, column mapping ui with type detection, preview of the first 20 rows, then commit. the mapping step is the hard part visually", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "add etag support to the collection endpoints, weak etags off updated_at max, and honor if-none-match", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "our github workflows have the same 30 lines of setup copy pasted 9 times. make it a composite action", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.4, "slice": "core", "lang": "en"}
|
|
{"prompt": "explain what the `--legacy-peer-deps` in our install step is papering over, and whether we can drop it", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "documentation for the event schema — every event type, its fields, when it fires, and which are guaranteed vs best effort", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "haptics on the dial control, subtle tick at each 5 degree stop, and it should respect the system haptics setting", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.4, "slice": "core", "lang": "en"}
|
|
{"prompt": "plan the observability work: traces we don't have, the 4 golden signals per service, and what we do about the fact that half our services don't propagate context", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "chown in the entrypoint takes 90 seconds on a big volume, we should only do it if needed", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.3, "slice": "core", "lang": "en"}
|
|
{"prompt": "server returns 500 on this payload and the error says 'invalid input syntax for type json' but the payload is valid json, i validated it three ways:\n\nPOST /v1/ingest\n{\n \"source\": \"crm\",\n \"records\": [\n {\"id\": \"a1\", \"attrs\": {\"note\": \"customer said \\\"maybe\\\" on renewal\"}},\n {\"id\": \"a2\", \"attrs\": {\"note\": \"path is C:\\\\Users\\\\temp\"}}\n ]\n}\n\nresponse:\n{\"error\":\"internal\",\"request_id\":\"req_8812f\"}\n\nserver log:\nERROR ingest: pq: invalid input syntax for type json\nDETAIL: Token \"Users\" is invalid.\nCONTEXT: JSON data, line 1: ...\"note\": \"path is C:\\Users...\n\nwe're passing the attrs through to a jsonb column. something is unescaping on the way", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "pasted-context", "lang": "en"}
|
|
{"prompt": "design the permissions model for shared folders and then build the check function, i'll do the endpoints", "purpose": "planning", "secondary": "backendImpl", "mixed": true, "difficulty": 0.8, "slice": "mixed", "lang": "en"}
|
|
{"prompt": "review the migration i wrote and if it's wrong just fix it", "purpose": "review", "secondary": "quickFix", "mixed": true, "difficulty": 0.5, "slice": "mixed", "lang": "en"}
|
|
{"prompt": "eh, next thing", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.3, "slice": "vague-eval", "lang": "en"}
|
|
{"prompt": "i want an rfc for how we'd expose usage-based billing. metering, aggregation windows, what we do about clock skew across regions, and how a customer disputes a number", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.9, "slice": "core", "lang": "en"}
|
|
{"prompt": "kafka consumer in go for the enrichment topic, manual commit after the sink succeeds, and a dlq topic for poison messages with the original headers preserved", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "the dashboard needs a date range picker — presets plus custom, and it should sync to the url so people can share a view", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "sentry sample rate is 1.0 in prod, take it to 0.2", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.1, "slice": "core", "lang": "en"}
|
|
{"prompt": "our api client is generated but we've hand edited it 6 times. regenerate properly and move the hand edits into wrapper functions so the next regen doesn't clobber them", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "ci passes, prod breaks. the difference is the timezone on the runners i think but the failure is in the date math and i can't pin it down", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "read the pr and tell me if the new index is actually going to be used by the query it's meant for", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "postmortem writeup for the double-charge incident. timeline, root cause, what we're changing. i'll fill in the timestamps", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "toolbar buttons in the ipad app are too close to the edge in landscape, and the popover anchors to the wrong point", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.4, "slice": "core", "lang": "en"}
|
|
{"prompt": "implement graceful shutdown properly — stop accepting, drain in flight with a 30s cap, close the db pool last, and log what got dropped", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "the same regex for parsing our resource ids appears in 11 files. one function, one place, and add the tests that pin the edge cases we've hit", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "faites-moi une revue de la couche de cache, surtout l'invalidation. pas de modifications, juste votre avis", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "fr"}
|
|
{"prompt": "readme badges are broken and the install instructions reference a package name we changed", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.2, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "spike out how we'd do a/b tests on the ios app without a server round trip per assignment", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "the linter wants explicit return types on 40 exported functions, add them", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.3, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "search returns nothing for queries with an apostrophe. works for everything else", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "give me the rundown on what happens during our blue/green deploy and where a request can get dropped, i'm writing a doc after this", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "chart component for the metrics page, line + area, 3 series max, live updating without a full re-render, and a crosshair tooltip", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "no", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.3, "slice": "vague-eval", "lang": "en"}
|
|
{"prompt": "we're going to need a plan for the data residency requirement. eu customers' data stays in eu, including backups and logs, and our current architecture assumes one region", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.9, "slice": "core", "lang": "en"}
|
|
{"prompt": "implement the diff algorithm for our document sync — myers is probably fine, operating on paragraph blocks not characters", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "avatar stack component, overlapping circles, +N overflow chip, tooltip listing everyone on hover", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.4, "slice": "core", "lang": "en"}
|
|
{"prompt": "readonly db user has write grants, revoke them", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.2, "slice": "core", "lang": "en"}
|
|
{"prompt": "our components import from '../../../lib/x' all over. set up path aliases and rewrite the imports, no other changes", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.4, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "kubectl logs show the pod handling the request but the trace has a gap of 4 seconds with no span. what could be eating that time", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "explain the difference between our `sync` and `reconcile` functions, they both seem to do the same thing and i don't want to delete the wrong one", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "docs page comparing our three storage backends for users choosing one — honest about tradeoffs, not marketing", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "in-app purchase flow with storekit 2, restore purchases, and handle the case where the receipt validates but our server doesn't know about the user yet", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "the modal's backdrop click closes it even when a form is dirty, gate that behind a confirm", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.4, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "move the 12 constants duplicated between the client and the server into a shared package, single source", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.4, "slice": "core", "lang": "en"}
|
|
{"prompt": "how bad is it that we're storing the api keys hashed with sha256 and no salt? i know it's not great, tell me how not-great and what the actual exposure is", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "our error messages say things like 'invalid input'. rewrite the user-facing copy for the 20 most common ones to say what to actually do", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.4, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "map out the steps to get our terraform out of one giant state file. 400 resources, and we can't have an outage during the moves", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "the go.sum has a stale entry after the replace directive, tidy it", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.1, "slice": "core", "lang": "en"}
|
|
{"prompt": "gh actions annotation from the failing job, this is blocking three PRs:\n\nError: The process '/usr/bin/git' failed with exit code 128\n ##[error]fatal: could not read Username for 'https://github.com': No such device or address\n at ExecState._setResult (/home/runner/work/_actions/actions/checkout/v4/dist/index.js:2417:25)\n at ExecState.CheckComplete (/home/runner/work/_actions/actions/checkout/v4/dist/index.js:2400:18)\n\nWarning: The `set-output` command is deprecated\nWarning: Node.js 16 actions are deprecated\n\nRun actions/cache@v4\n Cache not found for input keys: node-cache-Linux-x64-8f21a...\n\nit only fails on PRs from forks. worked last week. we did rotate a PAT on tuesday but that shouldn't matter for forks right", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "pasted-context", "lang": "en"}
|
|
{"prompt": "toast when a background upload finishes, tappable to jump to the file, and it shouldn't cover the tab bar", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.4, "slice": "core", "lang": "en"}
|
|
{"prompt": "build the reconciliation report: for every order in our db compare against the provider's settlement file, output the mismatches as csv with a reason code", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "our validation lives in three layers and they disagree on what a valid slug is. pick the strictest, put it in one place, everything calls that", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "what do these two goroutines actually synchronize on, i can't tell if the channel or the mutex is doing the work here", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "adr for choosing sqlite over postgres for the desktop app's local store. i want the reasoning captured before people forget", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.4, "slice": "core", "lang": "en"}
|
|
{"prompt": "plan for handling schema changes in the mobile app's local db when users skip 4 versions", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "dark mode in the ios app inverts the brand color in one place, ColorTheme.accent i think", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.2, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "app store connect rejected the build for a privacy manifest thing, i don't understand what it wants:\n\nITMS-91053: Missing API declaration - Your app's code in the \"AcmeNotes\" file references one or more APIs that require reasons, including the following API categories: NSPrivacyAccessedAPICategoryUserDefaults. While no action is required at this time, starting May 1, 2024, when you upload a new app or app update, you must include a NSPrivacyAccessedAPITypes array in your app's privacy manifest.\n\nITMS-91061: Missing privacy manifest - Your app includes \"Frameworks/SentrySwift.framework/SentrySwift\", which is a third-party SDK that requires a privacy manifest.\n\nwe do have a PrivacyInfo.xcprivacy, i added it last release", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "pasted-context", "lang": "en"}
|
|
{"prompt": "explain how our session invalidation works on password change, and whether it covers the mobile refresh tokens", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "write the migration guide for the plugin api v2 and mark which v1 methods are gone versus renamed", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "settings screen on android: preference screens with categories, a slider for sync frequency, and the switch states have to survive process death", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "cursor based iterator over the s3 inventory manifest, python, lazily fetching pages, and resumable from a saved cursor", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "sketch how we'd support undo across the whole editor. command pattern probably, but the tricky bit is collaborative edits landing in the middle of your undo stack", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.9, "slice": "core", "lang": "en"}
|
|
{"prompt": "unwrap the double negative in shouldSkipValidation, it's `!isNotDisabled` and nobody can read it", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.2, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "requests to /reports time out at 30s for exactly one customer. their dataset isn't the biggest. i've run their query manually and it takes 400ms", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "put together the internal doc on how our feature flags work and who can flip what, then i want to make the ui match it", "purpose": "writing", "secondary": "frontendImpl", "mixed": true, "difficulty": 0.5, "slice": "mixed", "lang": "en"}
|
|
{"prompt": "work out an approach for the batch export queue and then build the worker side", "purpose": "planning", "secondary": "backendImpl", "mixed": true, "difficulty": 0.7, "slice": "mixed", "lang": "en"}
|
|
{"prompt": "keep going where we left off on the sync engine", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "vague-eval", "lang": "en"}
|
|
{"prompt": "ok stream of thought — we have four services that all need the org tree, one of them caches it for 5 min, one hits the db every request, one has a copy in redis that nobody refreshes. i don't want to patch this, i want the actual strategy for owning that data in one place. write it up", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "implement the tar-like archive format we specced, streaming write, checksums per entry, and a verify subcommand", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "the file tree in the sidebar needs multi-select with shift and cmd, and a right click menu that respects the selection", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "the CORS preflight cache is 5s, bump to an hour", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.1, "slice": "core", "lang": "en"}
|
|
{"prompt": "convert the class components in src/legacy to function components with hooks, one directory at a time, snapshots must not change", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "duplicate notifications, two per event, started after we scaled the worker deployment to 2 replicas. i thought the queue handled that", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "is our idempotency actually idempotent if the first request is still in flight when the retry arrives", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "docs for the CLI's config file — every key, defaults, and which ones can be set by env var. table format", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.4, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "add prometheus metrics to the go service: request duration histogram by route, in flight gauge, and a counter for each error class", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "the print stylesheet is nonexistent, invoices print with the nav bar and dark backgrounds", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.4, "slice": "core", "lang": "en"}
|
|
{"prompt": "delete the commented out code in payments/legacy_gateway.py, it's been there since 2023", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.1, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "collapse our 5 environment config files into one with overlays, same effective values per env — diff them before and after to prove it", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "check the new websocket reconnect logic for me, specifically whether we can end up with two open sockets", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "we promised a status page. write the incident communication templates — investigating, identified, monitoring, resolved — with placeholders", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.3, "slice": "core", "lang": "en"}
|
|
{"prompt": "outline the plan for supporting webhooks with customer-provided mTLS certs, including how we store and rotate them", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "one of our cron jobs runs twice on the last sunday in october. i'm pretty sure i know why but the fix isn't obvious to me", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "swipe actions on the list rows — archive on left, delete on right with a confirm, and undo via a snackbar", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "das Feature-Flag `new_billing_ui` ist nur noch an einer Stelle relevant, räum die alten Verzweigungen weg ohne das Verhalten zu ändern", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.4, "slice": "boundary", "lang": "de"}
|
|
{"prompt": "so uh", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.3, "slice": "vague-eval", "lang": "en"}
|