topbot.fyi

The 69 build tasks of the harm check.

Most are features re-implemented from their pull-request description inside the repository's parent commit, with the commit's own tests held out; the rest are greenfield projects built from a written specification. Scores are the mean share of hidden tests passed, stock Claude Code and with the setup.

TaskSourceStackHidden testsStockSetup
Lightning-AI__pytorch-lightning-eea74433ceLightning-AI/pytorch-lightningpy160.750.69
PyCQA__pyflakes-d539458de5PyCQA/pyflakespy101.001.00
aio-libs__yarl-09e500de4baio-libs/yarlpy121.001.00
alecthomas__kong-4e565e8d77alecthomas/konggo11.001.00
gdamore__tcell-d57374f097gdamore/tcellgo21.001.00
go-resty__resty-29010be3b2go-resty/restygo11.001.00
golang-jwt__jwt-73c870b18egolang-jwt/jwtgo11.001.00
h3js__rou3-01bb393e2bh3js/rou3ts41.001.00
h3js__rou3-79480e642fh3js/rou3ts11.001.00
h3js__rou3-835332aca0h3js/rou3ts121.001.00
h3js__srvx-555a7932aeh3js/srvxts31.001.00
h3js__srvx-6b3b2910aeh3js/srvxts460.890.89
h3js__srvx-e0934021d4h3js/srvxts181.001.00
hgrecco__pint-bbc6b4c600hgrecco/pintpy10.000.00
joke2k__faker-4ca9116ec2joke2k/fakerpy21.001.00
junegunn__fzf-f779e6a4dfjunegunn/fzfgo21.001.00
labstack__echo-abbb92edbblabstack/echogo21.001.00
mark3labs__mcp-go-e9a7badccemark3labs/mcp-gogo71.000.50
ml-explore__mlx-lm-e9308d7b08ml-explore/mlx-lmpy51.001.00
modelcontextprotocol__go-sdk-3f43bcf012modelcontextprotocol/go-sdkgo11.001.00
mswjs__interceptors-4c01ff252emswjs/interceptorsts61.001.00
mswjs__interceptors-7f1b37ca80mswjs/interceptorsts60.670.67
openai__openai-go-7e4e1b1912openai/openai-gogo20.500.50
openai__openai-go-a29869f512openai/openai-gogo100.600.60
pypa__packaging-23669b1b69pypa/packagingpy20.500.50
pypa__packaging-58c6cd700bpypa/packagingpy540.890.83
python-attrs__attrs-5aa76a4450python-attrs/attrspy31.001.00
slice-go-cli-request-latency-reportgreenfield project from a written specgo131.001.00
slice-go-cli-sample-crate-plannergreenfield project from a written specgo401.001.00
slice-go-web-gallery-cue-cataloggreenfield project from a written specgo351.001.00
slice-go-web-signed-event-gatewaygreenfield project from a written specgo121.001.00
slice-go-worker-digest-packet-workergreenfield project from a written specgo361.001.00
slice-go-worker-revision-rebuild-workergreenfield project from a written specgo111.001.00
slice-py-cli-inventory-journal-foldgreenfield project from a written specpy121.001.00
slice-py-cli-prefix-policy-forgegreenfield project from a written specpy400.950.95
slice-py-web-buffered-dock-boardgreenfield project from a written specpy401.001.00
slice-py-web-gallery-caption-pressgreenfield project from a written specpy400.970.97
slice-py-web-harbor-berth-ledgergreenfield project from a written specpy381.001.00
slice-py-web-incident-timeline-pagesgreenfield project from a written specpy121.001.00
slice-py-worker-checkpointed-maintenance-workergreenfield project from a written specpy121.000.92
slice-py-worker-porchlight-couriergreenfield project from a written specpy491.001.00
slice-ts-cli-asset-revision-scoutgreenfield project from a written spects381.000.95
slice-ts-cli-catalog-placeholder-auditgreenfield project from a written spects121.001.00
slice-ts-cli-deploy-wave-plannergreenfield project from a written spects121.001.00
slice-ts-cli-env-overlay-plangreenfield project from a written spects121.001.00
slice-ts-cli-redirect-chain-auditorgreenfield project from a written spects101.001.00
slice-ts-cli-support-log-scrubbergreenfield project from a written spects121.001.00
slice-ts-web-apiary-inspection-apigreenfield project from a written spects501.001.00
slice-ts-web-artifact-manifest-registrygreenfield project from a written spects131.001.00
slice-ts-web-atomic-lane-layoutgreenfield project from a written spects121.001.00
slice-ts-web-durable-dispatch-deskgreenfield project from a written spects401.001.00
slice-ts-web-leased-review-queuegreenfield project from a written spects121.001.00
slice-ts-web-metered-usage-ledgergreenfield project from a written spects421.001.00
slice-ts-web-restoration-queue-apigreenfield project from a written spects351.001.00
slice-ts-web-transient-note-vaultgreenfield project from a written spects411.001.00
slice-ts-web-weekly-availability-apigreenfield project from a written spects121.001.00
slice-ts-worker-attachment-scan-workergreenfield project from a written spects121.001.00
slice-ts-worker-caption-proof-conveyorgreenfield project from a written spects401.001.00
slice-ts-worker-mirror-outbox-workergreenfield project from a written spects121.001.00
slice-ts-worker-receipt-gated-workflowsgreenfield project from a written spects121.001.00
slice-ts-worker-tenant-export-dispatchergreenfield project from a written spects131.001.00
stanfordnlp__dspy-33aaa19e03stanfordnlp/dspypy80.880.81
stanfordnlp__dspy-d58c3007ebstanfordnlp/dspypy120.330.33
stanfordnlp__dspy-e8ec009c6bstanfordnlp/dspypy310.240.34
tobymao__sqlglot-da9a9c9664tobymao/sqlglotpy180.670.64
tobymao__sqlglot-e3ed74860ftobymao/sqlglotpy20.500.50
tox-dev__filelock-126cab7456tox-dev/filelockpy310.740.74
tox-dev__platformdirs-dfaeabf921tox-dev/platformdirspy70.860.86
yuin__goldmark-ee62f0eaf9yuin/goldmarkgo20.000.50