// the find
DataDog/integrations-core
Core integrations of the Datadog Agent
The official source for all integrations that ship with the Datadog Agent — roughly 300+ checks covering everything from Active Directory to Zookeeper. If you run Datadog and want to understand how their agent checks work, or want to contribute a fix to one, this is the repo. Not useful outside the Datadog ecosystem.
Each integration follows a strict, consistent structure (manifest.json, spec.yaml, config_models/, metadata.csv) enforced by ddev tooling — you can read any check and immediately know where everything is. The supply-chain hygiene is serious: in-toto provenance links, pinned actions validation, Chainguard OIDC token workflows, and a dependency wheel promotion gate that prevents unpinned deps from shipping. The ddev CLI handles environment setup, test orchestration, and release management, which means the contributor workflow is genuinely reproducible across 300+ packages without per-check snowflakes. Tests run against real services via Docker, not mocks — the CI scripts in .ddev/ci/scripts/ handle service-specific setup like Kerberos installs and sysctl tweaks.
The low star count (1129) relative to fork count (1545) tells you this is mostly used by people who have to contribute fixes, not people who choose to study it — the code is operational glue, not something you'd read to learn patterns. Contributing a new check means internalizing a large amount of Datadog-specific tooling (ddev, spec.yaml schemas, config_models code generation) that has no value outside this repo. The repo is a monorepo of ~300 packages with a single CI matrix, so a flaky test in one obscure integration can block unrelated work — there are dedicated workflows just for tracking flaky tests on Windows, which signals this is an ongoing pain. Documentation for the check SDK lives on datadoghq.dev, not here, so the repo is hard to navigate without already knowing where to look.