azsdk-common-generate-sdk-pipeline
Skill135 starsChanged 2 months agoMITView on GitHub
Run the Azure SDK generation pipeline for a release plan and create the generated SDK pull requests, for one language or for all languages. **UTILITY SKILL**. USE FOR: "run SDK generation for all languages", "generate SDK for release plan <id>", "generate SDK for release <id>", "pipeline SDK generation", "generate SDK without a local clone", "create SDK pull requests". DO NOT USE FOR: generating a single SDK locally from a local clone (use azsdk-common-generate-sdk-locally), releasing/publishing an already-generated package (use azsdk-common-sdk-release), API design review. INVOKES: azure-sdk-mcp:azsdk_get_release_plan, azure-sdk-mcp:azsdk_run_generate_sdk, azure-sdk-mcp:azsdk_get_sdk_pull_request_link.
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 3 are distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 17
USE FOR: run SDK generation for all languages, generate SDK for a release plan, generate SDK for a release ID, pipeline SDK generation, …
- Distinctive: line 18
WHEN: "run SDK generation, run SDK generation for all languages for release plan <id>", "generate SDK for release plan <id>", "generate SDK …
- Very specific: line 19
DO NOT USE FOR: generating a single SDK locally from a local clone (use azsdk-common-generate-sdk-locally).
- Very specific: line 53
If the agent tries azsdk_release_sdk, get-artifact, or get-service-details tools instead of generating, redirect it to …
- Very specific: line 54
If the correct tool is not yet available in the session, activate/load the azure-sdk-mcp TypeSpec SDK toolset first, then call …
- Distinctive: line 55
If generation fails with a missing SDK details error, ensure the release plan has SDK details populated for the language before re-running.
≈ “Common” is an estimate: at least 1 in 5 of the 308,232 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Generate SDK via Pipeline
This skill runs the Azure SDK generation pipeline for a release plan and produces the generated SDK pull request(s). It is the correct workflow when the user asks to generate SDKs for all languages, to generate for a release plan / release ID, or to generate without a local clone — none of which the local-generation workflow handles.
Triggers
USE FOR: run SDK generation for all languages, generate SDK for a release plan, generate SDK for a release ID, pipeline SDK generation, generate SDK without a local clone, create SDK pull requests
WHEN: "run SDK generation, run SDK generation for all languages for release plan <id>", "generate SDK for release plan <id>", "generate SDK for release <id>", "pipeline SDK generation", "generate SDK without a local clone"
DO NOT USE FOR: generating a single SDK locally from a local clone (use azsdk-common-generate-sdk-locally); releasing or publishing an already-generated package (use azsdk-common-sdk-release); API design review
Rules
- Always call
azure-sdk-mcp:azsdk_run_generate_sdkto generate. Do not useazure-sdk-mcp:azsdk_release_sdk(that publishes an already-generated package, it does not generate) orazure-sdk-mcp:azsdk_get_sdk_pull_request_link/azure-sdk-mcp:azsdk_get_pull_request(those only retrieve links) to generate an SDK. azure-sdk-mcp:azsdk_run_generate_sdkgenerates one language per call. To generate for all languages, call it once per language the release plan targets.- A release plan work item ID (or release plan ID) is required.
azsdk_run_generate_sdkalso requires the TypeSpec project path and the SDK release type (betaorstable) as explicit inputs, and it validates them — it does not read them from the release plan. First callazure-sdk-mcp:azsdk_get_release_planto obtain the TypeSpec project path, SDK release type, and target languages, then pass those into each generation call along with the work item ID. - Generation uses the release plan’s saved spec commit when present. Plans without a saved commit keep the existing branch/PR behavior.
- Requires the
azure-sdk-mcpserver; there is no CLI fallback for the pipeline generation workflow. - Private Preview release plans cannot generate SDKs via the pipeline — only the API spec PR needs to merge. If needed for validation, direct the user to generate locally via
azsdk-common-generate-sdk-locally.
MCP Tools
| Tool | Purpose |
azure-sdk-mcp:azsdk_get_release_plan | Fetch the release plan to determine target languages / details |
azure-sdk-mcp:azsdk_run_generate_sdk | Run the generation pipeline (once per language) |
azure-sdk-mcp:azsdk_get_sdk_pull_request_link | Retrieve the generated SDK pull request link after generation |
Steps
- Collect release plan — Get the release plan work item ID (or release plan ID) from the user, then call
azure-sdk-mcp:azsdk_get_release_planto fetch it. - Determine languages — If the user asked for "all languages", determine the languages the release plan targets (e.g. via
azure-sdk-mcp:azsdk_get_release_plan). Otherwise use the single language requested. - Generate per language — For each target language, run
azure-sdk-mcp:azsdk_run_generate_sdkpassing the TypeSpec project path, the SDK release type (betaorstable), and the language taken from the release plan, plus the release plan work item ID. These inputs are required and validated — an invalid TypeSpec project path, or a release type other thanbeta/stable, fails the run. - Report pull requests — After each run completes, retrieve and show the generated SDK pull request link with
azure-sdk-mcp:azsdk_get_sdk_pull_request_link.
Examples
- "Run SDK generation for all languages for release 12345"
- "Generate the SDK for release plan 12345"
- "Generate the Python and .NET SDKs for release 12345 using the pipeline"
Troubleshooting
- If the agent tries
azsdk_release_sdk, get-artifact, or get-service-details tools instead of generating, redirect it toazure-sdk-mcp:azsdk_run_generate_sdk— that is the only tool that runs the generation pipeline. - If the correct tool is not yet available in the session, activate/load the
azure-sdk-mcpTypeSpec SDK toolset first, then callazure-sdk-mcp:azsdk_run_generate_sdk. - If generation fails with a missing SDK details error, ensure the release plan has SDK details populated for the language before re-running.
- Requires the
azure-sdk-mcpserver. No CLI fallback — prompt the user to configure MCP if unavailable.
In this repository43 other files it gives its agents.
- packages/python-packages/apiview-copilot/AGENTS.mdAGENTS.md
- .github/copilot-instructions.mdCopilot instructions
- azsdk-common-apiview-feedback-resolutionSkill
- azsdk-common-generate-sdk-locallySkill
- azsdk-common-pipeline-analysisSkill
- azsdk-common-pipeline-fixerSkill
- azsdk-common-prepare-release-planSkill
- azsdk-common-sdk-breaking-changeSkill
- azsdk-common-sdk-releaseSkill
- azsdk-common-typespec-namingSkill
- azure-typespec-assessmentSkill
- azure-typespec-authorSkill
- code-reviewSkill
- eval-authoring-skillSkill
- eval-authoring-toolSkill
- eval-authoring-workflowSkill
- markdown-token-optimizerSkill
- sdk-ai-bot-eval-datasetSkill
- sdk-ai-bot-run-evaluationSkill
- skill-authoringSkill
- add-endpointSkill
- audit-filtersSkill
- dev-setupSkill
- kb-searchSkill
- report-active-reviewsSkill
- report-apiview-metricsSkill
- report-architect-commentsSkill
- report-feedbackSkill
- report-memorySkill
- report-revisionsSkill
- run-evalsSkill
- run-metricsSkill
- update-guidelinesSkill
- apiview-feedback-resolutionSkill
- azure-typespec-authorSkill
- generate-sdk-locallySkill
- azsdk-common-live-and-recorded-testsSkill
- markdown-token-optimizerSkill
- pipeline-troubleshootingSkill
- azsdk-common-prepare-release-planSkill
- azsdk-common-sdk-releaseSkill
- skill-authoringSkill
- pr-review-helpSkill
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
Reviews can't be read right now.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

