CtrlK
BlogDocsLog inGet started
Tessl Logo

mcp-builder

Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP), Node/TypeScript (MCP SDK), or C#/.NET (Microsoft MCP SDK).

79

2.50x
Quality

68%

Does it follow best practices?

Impact

100%

2.50x

Average score across 3 eval scenarios

SecuritybySnyk

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./.github/skills/mcp-builder/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

50%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is a well-structured four-phase guide with concrete commands and good navigation design, but it is held back by re-stated general knowledge, a lack of inline executable code, weak validation feedback loops, and reference links to files that are absent from the bundle.

Suggestions

Trim general-principle restatements (DRY, async/await, descriptive naming) to keep only MCP-specific guidance, improving token efficiency.

Add explicit validate→fix→retry feedback loops in Phase 3 (e.g., gate testing on successful compilation with a re-run-on-failure step) to raise workflow clarity.

Either include the missing reference/*.md files in the bundle or remove/inline the broken links so the progressive-disclosure structure is intact.

DimensionReasoningScore

Conciseness

Mostly efficient tables and structure, but several sections restate general software-engineering knowledge Claude already has ("async/await for I/O operations", "No duplicated code (DRY principle)", "Clear, descriptive tool names help agents find the right tools quickly") that could be trimmed.

3 / 5

Actionability

Provides concrete commands (npx @modelcontextprotocol/inspector, npm run build, python -m py_compile) and specific SDK URLs and annotation flags, but defers all executable code to reference files with no inline copy-paste examples, leaving a gap.

3 / 5

Workflow Clarity

The four-phase sequence (Research → Implementation → Review/Test → Evaluations) is clear with build/test steps, but validation is presented as bullet tasks rather than explicit pass/fail gates and there is no validate→fix→retry feedback loop.

3 / 5

Progressive Disclosure

The reference structure is well-designed and clearly signaled on paper (emoji-labeled one-level-deep links to per-topic reference/*.md files organized by load timing), but none of the referenced reference files actually exist in the bundle, leaving the overview pointing to a missing documentation library.

3 / 5

Total

12

/

20

Passed

Description

87%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, third-person description that clearly states both what the skill does and when to use it, with concrete SDK/language specifics and natural trigger phrasing. Minor improvements could add user-synonym triggers and a fuller enumeration of tool-authoring operations.

DimensionReasoningScore

Specificity

Names the domain (MCP servers) plus several concrete specifics — "creating high-quality MCP servers", "integrate external APIs or services", and three named SDKs (FastMCP, MCP SDK, Microsoft MCP SDK) — with only minor gaps in enumerated operations.

4 / 5

Completeness

Explicitly answers both what ("Guide for creating high-quality MCP servers that enable LLMs to interact with external services through well-designed tools") and when ("Use when building MCP servers to integrate external APIs or services...") with concrete trigger phrases.

5 / 5

Trigger Term Quality

Good natural-trigger coverage ("Use when building MCP servers", "integrate external APIs or services") plus language/SDK keywords, though it misses common user phrasings like "connect my API to Claude" or "expose tools".

4 / 5

Distinctiveness Conflict Risk

MCP/Model Context Protocol with named SDKs is a clearly distinct niche with minimal overlap risk; the MCP-specific framing keeps it from triggering for unrelated skills.

5 / 5

Total

18

/

20

Passed

Validation

93%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

relative_links

Relative link issues: 17 missing

Warning

Total

15

/

16

Passed

Repository
microsoft/skills
Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.