Agent skill · data analytics · pytorch
ci-metrics
Query PyTorch CI, GitHub Actions, HUD, Grafana, and infrastructure metrics. Use when users ask about CI duration, job failures, queue times, workflow trends, runner health, dashboard data, or PyTorch infrastructure metrics.
Why this skill is useful
Provides specific commands and workflows for querying CI metrics from Grafana and GitHub Actions that the AI wouldn't reliably generate on its own.
What it needs
Requires curl, gh installed locally. About 2k tokens when loaded. Last updated 2026-08-07. 102,253 stars on the source repository.
What this skill does
PyTorch CI Metrics PyTorch CI and infrastructure metrics are exposed through Grafana. Use .claude/skills/ci-metrics/gcx-wrapper.sh for all Grafana access; it configures the PyTorch Grafana server, context, and authentication. Only users with write permission to the repo have access to Grafana. The authentication only provides read only access. Requirements The wrapper needs these tools on PATH: gh - fetches the Grafana token and must be authenticated; if not, run gh auth login --hostname github.com --git-protocol ssh --web. curl - downloads gcx and fetches the token from HUD. On first use the wrapper downloads a pinned, checksum-verified gcx binary into a private cache (~/.cache/pytorch-ci-metrics/) and authenticates automatically. Nothing is installed on your PATH. If a tool is missing or gh is not authenticated, it exits with an error describing what to fix. Datasources Get the list of datasources available: The data contains metrics for many repos owned by the PyTorch repo. When possible, restrict queries to just the pytorch/pytorch repository. CI and Test Run Data CI and test run data are stored in grafana-clickhouse-datasource. List all the available tables: Important dataset: GitHub webhook data Database: default Note: the default database also contains other non-webhook related tables. Learn more about the event and payload: https://docs.github.com/en/webhooks/webhook-events-and-payloads Tests Database: tests tests.alltestruns - contains every test run. This is an extremely large table, so be considerate with filtering and timing. Do not use tests.testruns3 as it contains partial data only. To get additional guidance on common queries, clone https://github.com/pytorch/test-infra into a temporary directory and read the torchci folder. …
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills pytorch/ci-metrics