OSDU Claude Plugins
An agentic system for OSDU, built for Claude Code
A Claude Code marketplace that bundles specialized agents, skills, and live platform access into two installable plugins for infrastructure automation, platform analytics, QA testing, and knowledge management across OSDU Community Implementation environments on Azure.
Built for OSDU maintainers, platform engineers, QA contributors, and community operators.
| Plugin | Components | What you use it for |
|---|
| osdu | osdu:osdu agent, 21 skills | Builds, dependencies, MRs, tests, live platform queries, knowledge management |
| cimpl | cimpl:cimpl agent, 3 skills | Terraform, Helm, AKS, debugging, verification |
Install
Requires Claude Code.
/plugin marketplace add danielscholl/claude-osdu
/plugin install osdu@claude-osdu
/plugin install cimpl@claude-osdu
/reload-plugins
Use
> give me a morning briefing
> clone partition
> what pipelines are failing across OSDU?
> check the health of my environment
> run acceptance tests for partition
> review MR !845
> ship it
Plugins
osdu
Platform operations across 30+ OSDU GitLab services.
| Category | Skills |
|---|
| Knowledge | brain, briefing, learn, consolidate |
| Analytics | osdu-activity, osdu-engagement, osdu-quality |
| Build & Deps | maven, dependency-scan, build-runner, remediate |
| QA Testing | osdu-qa, acceptance-test |
| Git Workflow | send, mr-review, contribute, glab, fossa, maintainer |
| Workspace | clone, setup |
Agents: osdu:osdu (orchestrator), osdu:build-runner, osdu:qa-runner, osdu:qa-analyzer, osdu:qa-comparator, osdu:qa-reporter
MCP Server: osdu-mcp-server for live platform access (31 tools across search, storage, schema, entitlements, legal, partition)
cimpl
Azure infrastructure automation for OSDU Community Implementation deployments.
| Category | Skills |
|---|
| Infrastructure | iac, health, setup |
Agent: cimpl:cimpl (infrastructure specialist)
Testing
Uses the skilltest four-layer test framework.
make test # L1 + L2 + pytest (fast)
make lint # L1: Structure validation
make unit # L2: Trigger eval dry-run
make integration P=osdu # L3: Multi-turn session tests
make benchmark P=osdu S=brain # L4: Skill value comparison
make report # Test inventory
Contributing
See CONTRIBUTING.md.
License
Apache License 2.0
danielscholl