Skip to content
Agentmart
Submit a tool

mcp-engineering

Build MCP servers with well-named tools, workflow-oriented coverage, and clear errors.

Skill by chnetajibc

IndexedMIT
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/chnetajibc-mcp-engineering

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Build MCP servers with well-named tools, workflow-oriented coverage, and clear errors. Use when exposing capabilities to agents via MCP. Pair with security/agent-security for the audit side.

Use cases: testing, security, productivity, agent-building

Reviews

Write a review

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/mcp-engineering

What's inside

Skill name
mcp-engineering
Description
Build MCP servers with well-named tools, workflow-oriented coverage, and clear errors. Use when exposing capabilities to agents via MCP. Pair with security/agent-security for the audit side.
License
MIT
Instructions
26 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

# mcp-engineering

Build the tool the agent wishes existed: workflow-shaped, narrowly permissioned, loudly correct on misuse.

## Activate when

- Creating or extending an MCP server (Python FastMCP or TypeScript).
- Deciding tool boundaries, naming, or error formats.

## Do NOT activate for

- Auditing an existing MCP server for security (see security/agent-security).

## Procedure

1. Scope: one tool per user-meaningful workflow step, not per internal function; prefer fewer, composable tools over exhaustive coverage.
2. Naming: `verb-object` names, typed inputs, documented preconditions; errors state what was wrong, what was expected, and the safe retry.
3. Permissions: least privilege per tool; destructive tools require confirmation params and log intent.
4. Versioning: additive changes first; breaking tool-shape changes get a version bump + migration note.
5. Evaluate: scripted happy-path + misuse + adversarial-input runs before shipping.
6. Verify: run the eval script, exercise tools from a real client, and confirm logs show who called what.

## References
- Related: `../agent-architecture/`, `../../security/agent-security/`.

From chnetajibc/skill-repo, MIT.

More from chnetajibc/skill-repo

Signals

Stars
0
Forks
0
Open issues
0
Last commit (this folder)
Sep 22, 2026 (1 month ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
Repository
github.com/chnetajibc/skill-repo
First seen
Oct 24, 2026
Last checked
Oct 24, 2026
Source
GitHubgithub-graphql

More from chnetajibc

Cite this page

mcp-engineering skill. Agent Mart. https://agentmart.dev/skills/chnetajibc-mcp-engineering (last updated 2026-10-24)