2.6k
GitHub Stars
2
Bundled Files
2 months ago
Catalog Refreshed
3 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill openclaw/skills --skill agentic-mcp-server-builder- _meta.json310 B
- SKILL.md910 B
Overview
This skill scaffolds minimal MCP server projects and generates baseline tool contract checks from a structured tool list. It produces a starter file map, contract summaries for each tool, and can materialize starter files when run outside dry-run mode. The goal is to accelerate MCP-ready server setup with clear input/output schemas and deterministic tool boundaries.
How this skill works
You provide a server name and a list of tools with short descriptions and schemas. The skill generates a scaffold file map, a contract summary for each tool, and optional starter files when not in dry-run. It also emits baseline contract checks that highlight required inputs, outputs, and determinism expectations for MCP compliance.
When to use it
- Bootstrapping a new MCP server with multiple tools
- Defining clear tool schemas and input/output contracts before coding business logic
- Creating a reproducible starter layout for team handoff or onboarding
- Validating that an existing project matches MCP contract expectations
- Generating lightweight artifacts for archival or backup of tool definitions
Best practices
- Keep each tool small and focused with explicit input and output schemas
- Use deterministic outputs where possible and document nondeterministic behavior
- Run in dry-run first to review file map and contract checks before writing files
- Include schema examples and edge-case test inputs in the generated contracts
- Maintain one source of truth for tool definitions to simplify updates and archives
Example use cases
- Define a chat-processing tool list and generate an MCP server scaffold to implement handlers
- Convert a set of Python utility functions into MCP-ready tools with validated input/output contracts
- Create an archived snapshot of all tool contracts and starter files for backup or audit
- Onboard new developers with a generated project layout and contract summary to speed implementation
FAQ
Yes. Run in dry-run mode to produce the scaffold map and contract summary without materializing files.
Does it enforce schemas automatically?
It generates baseline contract checks and schema suggestions; you should review and extend those checks to enforce domain-specific validation.