agentleFS
mgovedarovmgovedarov/mcp-vcf-orchestrator/skills/vcfa-operations/SKILL.md

vcfa-operations

Skill7 starsChanged 17 days agoApache-2.0View on GitHub

Run and operate VCF Automation Orchestrator, Service Broker, and Cloud Assembly through the mcp-vcf-orchestrator MCP server. Use when the user wants to run a workflow, troubleshoot a failed workflow execution or a stuck deployment, explore catalog items and deployments, manage day-2 actions, create or review blueprint templates, work with extensibility subscriptions and event topics, or discover what an environment offers. Encodes read-only-first discovery and confirm-before-write safety, and routes to the matching vcfa-* prompts.

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Treat it like a production operation: confirm target, inputs, and blast radius.

    line 36
  2. Very specific:

    When the server runs with VCFA_TARGET_PLATFORM=vra8 (vRA/vRO 8.12+, vIDM bearer-token auth), the whole vRO surface works, reads and writes …

    line 76

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

VCFA Operations

Operate and troubleshoot VCF Automation through the MCP server. Start with read-only discovery and only mutate after explicit confirmation. This skill orchestrates the server's prompts, resources, and tools rather than replacing them.

Core principles

  • Inspect before you act. Use the list-* and get-* tools and the execution log tools to understand state before remediating.
  • The mutating tools — run-workflow, run-workflow-and-wait, create-deployment, run-deployment-action, delete-deployment, create-template, delete-template, create-subscription, update-subscription, delete-subscription — require a confirm: true argument plus real user confirmation of the exact target and expected impact.
  • create-deployment provisions real infrastructure and is annotated destructive for that reason — the one additive create- tool that is. Both of its required arguments are opaque UUIDs that nothing else in the call identifies, so pass expectedCatalogItemName and expectedProjectName and let the handler verify the target against live metadata before it submits.
  • Running a workflow or a deployment day-2 action changes the live environment. Treat it like a production operation: confirm target, inputs, and blast radius.

Prefer the server prompts

  • vcfa-troubleshoot-deployment — inspect a deployment and its day-2 options before remediation.
  • vcfa-troubleshoot-workflow-execution — diagnose a failed execution using logs, stack, and workflow source.
  • vcfa-discover-capabilities — explore plugins, categories, workflows, actions, catalog items, deployments, templates, and subscriptions.
  • vcfa-collect-context-snapshot — persist reusable environment context.
  • vcfa-create-template and vcfa-review-template — discover conventions before drafting a blueprint, and review one for catalog readiness.
  • vcfa-integrate-workflow-template-subscription — plan workflow, template, catalog, deployment, and subscription integration.

Tool map by task

  • Execution: run-workflow, run-workflow-and-wait, get-workflow-execution, get-workflow-execution-logs, list-workflow-executions.
  • Deployments: list-deployments, get-deployment, list-deployment-requests, get-deployment-request, list-deployment-actions, run-deployment-action, create-deployment (provisions infrastructure; annotated destructive), delete-deployment.
  • Projects: list-projects, get-project (resolve projectId before project-scoped calls).
  • Catalog: list-catalog-items, get-catalog-item.
  • Templates: list-templates, get-template, create-template, delete-template.
  • Subscriptions and events: list-event-topics, list-subscriptions, get-subscription, create-subscription, update-subscription, delete-subscription.
  • Discovery support: list-workflows, list-actions, list-categories, list-plugins.

Live objects are also available as read-only resources, e.g. vcfa://deployments/{id}, vcfa://workflows/{id}, and vcfa://subscriptions/{id}.

vra8 mode caveat

When the server runs with VCFA_TARGET_PLATFORM=vra8 (vRA/vRO 8.12+, vIDM bearer-token auth), the whole vRO surface works, reads and writes alike, and the Automation-service surfaces — catalog, deployments, templates, projects, subscriptions, and event topics — can be read. Template and subscription writes work there too: creating or deleting a template, and creating, updating, or deleting a subscription. Deployment writes work there too — creating a deployment, deleting one, and running a day-2 action — verified on vRA 8.18 under VCFO-088, behind the same confirm gate and expected* guards as on VCF Automation. export-configuration-file is unsupported on every platform, not only there: no vRO tested serves a single configuration element as a .vsoconf, so the request answers 406. Route a configuration element through the project package instead. In vra8 mode prepare-artifact-promotion reports a configuration backup as skipped while still returning its report; elsewhere that backup attempts the export and surfaces the refusal.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

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.