patternfly / patternfly/patternfly-mcp

Allow MCP skills as MCP tools and prompts

Open
#186 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

PF Team
Dominant language
TypeScript
Stars
5
Forks
11
Avg merge
3d 5h
Merged PRs (30d)
17

Description

As a developer now that we've activated an MCP resource index for skills, see #185 , we can add the remaining activation facets for skills as MCP tools and prompts.

Acceptance criteria

  • Review the prompts and resources as plugins. Refine implementation

  • Lazy load skills and available artifacts once selected

  • Generate MCP prompt for skill - basic only. Advanced prompts may cause issues across available MCP clients

  • Generate a base skill as MCP tool.

    • The tool should provide progressive steps and require verification per step.
    • The tool should provide related artifacts ONLY at the time of their need as available MCP resources

Assumptions and questions

  • Discovery of skills should now include multiple paths

    • basic generated MCP prompts
    • MCP tool search/response results, (e.g. "[this skill appears related]" or "[these are all available skills]")
    • MCP indexed resource of skills
  • Need to determine the extent to which we maintain state of the steps. Do we handle this across sessions, at min maybe a "[stopped here, recommend you start over]"? "That should be left up to the model" would be the quickest implementation


Jira Issue: PF-4122

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by reviewing the existing prompt and resource plugin implementations, then trace skill discovery across generated prompts, tool search results, and indexed resources. Clarify lazy-loading and step-state expectations before implementation. Done means skills are available through basic MCP prompts and tools, with related artifacts exposed as resources when needed.

Written by the indexing model from the issue text.

Assessment

Tech stack
typescript
Domain
api, tooling
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Quiet
Clarity
Needs clarification
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.