Files
claude-skills-reference/eval/skills/mcp-server-builder.yaml
Leo 75fa9de2bb feat: add promptfoo eval pipeline for skill quality testing
- Add eval/ directory with 10 pilot skill eval configs
- Add GitHub Action (skill-eval.yml) for automated eval on PR
- Add generate-eval-config.py script for bootstrapping new evals
- Add reusable assertion helpers (skill-quality.js)
- Add eval README with setup and usage docs

Skills covered: copywriting, cto-advisor, seo-audit, content-strategy,
aws-solution-architect, agile-product-owner, senior-frontend,
senior-security, mcp-server-builder, launch-strategy

CI integration:
- Triggers on PR to dev when SKILL.md files change
- Detects which skills changed and runs only those evals
- Posts results as PR comments (non-blocking)
- Uploads full results as artifacts

No existing files modified.
2026-03-12 05:39:24 +01:00

42 lines
1.7 KiB
YAML

# Eval: mcp-server-builder
# Source: engineering/mcp-server-builder/SKILL.md
description: "Evaluate MCP server builder skill"
prompts:
- |
You are an expert AI assistant. You have the following skill loaded:
---BEGIN SKILL---
{{skill_content}}
---END SKILL---
Now complete this task: {{task}}
providers:
- id: anthropic:messages:claude-sonnet-4-6
config:
max_tokens: 4096
temperature: 0.7
tests:
- vars:
skill_content: file://../../engineering/mcp-server-builder/SKILL.md
task: "Build an MCP server in Python that exposes a 'search_github_repos' tool. The tool should take a query string and return top 5 repos with name, stars, and description. Use the GitHub REST API (no auth required for public search)."
assert:
- type: llm-rubric
value: "Output includes working Python code that follows MCP server patterns (tool registration, handler)"
- type: llm-rubric
value: "Code includes proper error handling for API failures"
- type: llm-rubric
value: "Tool definition includes proper input schema with type annotations"
- vars:
skill_content: file://../../engineering/mcp-server-builder/SKILL.md
task: "Design an MCP server architecture for a CRM system that exposes: list_contacts, get_contact, create_contact, search_contacts, and list_deals tools. Show the tool definitions and server structure."
assert:
- type: llm-rubric
value: "Response includes tool definitions with proper input/output schemas for all 5 tools"
- type: llm-rubric
value: "Architecture follows MCP best practices (proper transport, error handling, resource definitions)"