Skip to content
Back to skills

Create Mcp Integration

ASecurity

Adds an MCP server integration configuration to an existing plugin. NOT for scaffolding a brand-new plugin (use `create-plugin`) and NOT for Azure hosted agents (use `create-azure-agent`).

  • 7 stars
  • 0 votes
  • 0 copies
  • 2 views
  • Added September 2, 2026
ai-agentspythonbashazureapisecurity

Works with

  • api
  • mcp

Security analysis

A100/100

Pro scans all 20 files and shows the line behind each finding

Scanned October 3, 2026

npx -y skills add richfrem/agent-plugins-skills --skill create-mcp-integration --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Create Mcp Integration?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Create Mcp Integration
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/richfrem-create-mcp-integration-agent-plugins-skills/badge)](https://www.skillsdirectory.com/skills/richfrem-create-mcp-integration-agent-plugins-skills)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: create-mcp-integration
plugin: agent-scaffolders
description: >
  Adds an MCP server integration configuration to an existing plugin. NOT for scaffolding a brand-new plugin (use `create-plugin`) and NOT for Azure hosted agents (use `create-azure-agent`).
argument-hint: "[mcp-server-name or service]"
allowed-tools: Bash, Read, Write
---

# Create MCP Integration (create-mcp-integration)

Configures Model Context Protocol (MCP) server integrations within a plugin or workspace `.mcp.json`.

## Contents
- [Critical Constraints](#critical-constraints)
- [Quick start](#quick-start)
- [Workflow](#workflow)
- [Verification](#verification)
- [References](#references)

## Critical Constraints
- **Integration Scope**: Configuration of MCP servers only. For scaffolding new plugins, use `create-plugin`. For hosted cloud agents, use `create-azure-agent`.
- **Encrypted Transports**: Remote server endpoints must use `https://` or `wss://`; unencrypted plain HTTP/WS transports are forbidden.
- **Modern Transport Priority**: Prefer `streamable-http` or `stdio` transports over deprecated SSE implementations.
- **Credential Hygiene**: Store secrets strictly in environment variables; never embed raw API tokens or credentials inside `.mcp.json`.

## Quick start

```bash
python3 plugins/agent-scaffolders/scripts/scaffold.py \
  --type mcp \
  --name postgres-mcp \
  --path plugins/<plugin>
```

## Workflow

1. **Phase 1: Transport & Security Discovery**: Establish server archetype (`stdio`, `streamable-http`, `sse`), endpoint URL or binary command, and authentication mechanism.
2. **Phase 2: Tool Surface Definition**: Select target tool subsets, resource templates, and prompt templates to expose to the agent environment.
3. **Phase 3: Configuration Scaffolding**: Generate the `.mcp.json` server definition block with required environment variable placeholders and arguments.
4. **Phase 4: Schema & Connection Audit**: Validate JSON formatting and test connectivity using the environment MCP inspector.

## Verification

```bash
# Validate JSON formatting
python3 -c "import json; json.load(open('.mcp.json'))"

# Audit skill compliance
python3 plugins/agent-scaffolders/scripts/audit_skill.py plugins/agent-scaffolders/skills/create-mcp-integration --mode source
```

## References
- [fallback-tree.md](references/fallback-tree.md) - Fallback tree for scaffolding failures.
- [references/server-types.md](references/server-types.md) - MCP transport architecture standards.
- [references/authentication.md](references/authentication.md) - Safe credential and authentication patterns.
- [references/tool-usage.md](references/tool-usage.md) - Tool filtering and prompt exposure.
- [acceptance-criteria.md](references/acceptance-criteria.md) - Quality checklist for MCP integrations.

Files in this skill

  • SKILL.md1.4 KB
  • acceptance-criteria.md359 B
  • evals/evals.json881 B
  • evals/results.tsv417 B
  • fallback-tree.md1.4 KB
  • references/acceptance-criteria.md42 B
  • references/authentication.md37 B
  • references/examples/http-server.json48 B
  • references/examples/sse-server.json47 B
  • references/examples/stdio-server.json49 B
  • references/fallback-tree.md36 B
  • references/hitl-interaction-design.md46 B
  • references/pattern-decision-matrix.md46 B
  • references/patterns/action-forcing-output-with-deadline-attribution.md82 B
  • references/patterns/adversarial-objectivity-constraint.md69 B
  • references/patterns/anti-pattern-vaccination.md59 B
  • references/patterns/anti-symptom-triage.md54 B
  • references/patterns/artifact-embedded-execution-audit-trail.md74 B
  • references/patterns/artifact-generation-xss-compliance-gate.md74 B
  • references/patterns/artifact-lifecycle.md53 B

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.