CtrlK
BlogDocsLog inGet started
Tessl Logo

mcp-integration

This skill should be used when the user asks to "add MCP server", "integrate MCP", "configure MCP in plugin", "use .mcp.json", "set up Model Context Protocol", "connect external service", mentions "${CLAUDE_PLUGIN_ROOT} with MCP", or discusses MCP server types (SSE, stdio, HTTP, WebSocket). Provides comprehensive guidance for integrating Model Context Protocol servers into Claude Code plugins for external tool and service integration.

69

Quality

85%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

81%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.

A highly actionable, well-sequenced reference with concrete configs, an explicit workflow with validation and debugging loops, and real supporting bundle files. Its weaknesses are verbosity from repeated best-practice/checklist sections and generic error-handling advice, and a dangling examples/ reference alongside deep-dive content inlined in SKILL.md that belongs in the existing reference files.

Suggestions

Collapse the duplicated guidance: merge the final 'Best Practices' DO/DON'T block and 'Configuration Checklist' into the earlier 'Security Best Practices' and 'Testing MCP Integration' sections, and cut the 'Error Handling' bullets that restate what Claude already knows (e.g., 'Provide clear error messages').

Move the detailed per-server-type configuration and 'Authentication Patterns' sections into the existing references/server-types.md and references/authentication.md, keeping SKILL.md to one compact example per type plus the links.

Remove or actually create the 'examples/' directory — the body advertises stdio-server.json, sse-server.json, and http-server.json that are not present in the bundle.

DimensionReasoningScore

Conciseness

Most sections carry real, non-obvious payload (config schemas, tool naming format, lifecycle order), but there is substantial duplication and padding: security DO/DON'T guidance appears in "Security Best Practices" and again nearly verbatim in the final "Best Practices" section; env-var documentation is repeated four times; and "Authentication Patterns" re-shows SSE/HTTP configs already given under server types. "Error Handling" and "Performance Considerations" also dispense generic advice Claude already knows ("Provide clear error messages", "Validate inputs"). Not score 2, because the majority of tokens do convey plugin-specific facts Claude would not know; not score 4, because the repeated checklist/best-practice blocks and generic error-handling advice are clearly trimmable.

3 / 5

Actionability

Guidance is copy-paste ready throughout: complete JSON configs for every server type, the exact tool-name format (mcp__plugin_<plugin>_<server>__<tool>) with a worked example, concrete verification commands (claude --debug, /mcp), frontmatter allowed-tools snippets, and a 9-step implementation workflow. Common cases are covered by specific examples.

5 / 5

Workflow Clarity

The Implementation Workflow gives a clear 9-step sequence with test steps embedded (steps 5 and 8), backed by a dedicated Testing section with a numbered local-testing procedure, a validation checklist, and a Debugging section with issue→fix recovery paths for the three most common failures. Feedback loops and checkpoints are explicit, matching the top anchor.

5 / 5

Progressive Disclosure

Structure is good: three real reference files (verified to exist) are listed under "Reference Files" with one-line descriptions and are one level deep. It falls short of 5 because the 550-line body inlines substantial deep-dive material (full server-type and authentication details) that duplicates the stated purpose of the references files, and it points to an "examples/" directory (stdio-server.json, sse-server.json, http-server.json) that does not exist in the bundle — a dangling reference. Not 3, because references are clearly signaled and navigation is easy.

4 / 5

Total

17

/

20

Passed

Description

90%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 description with excellent, natural trigger coverage and a clear, explicit use-when clause. The only weakness is that the capability statement is a single abstract phrase ("comprehensive guidance") rather than a list of concrete actions, which slightly blurs what the skill actually does.

DimensionReasoningScore

Specificity

The capability statement is limited to one generic action — "Provides comprehensive guidance for integrating Model Context Protocol servers into Claude Code plugins for external tool and service integration" — which names the domain but does not enumerate concrete actions (e.g., configuring .mcp.json, setting up OAuth, pre-allowing tools). Score 4 would require several specific listed actions; score 2 would mean the domain itself was vague, but "MCP servers into Claude Code plugins" is concrete.

3 / 5

Completeness

Both halves are explicit: the "when" is a long list of concrete trigger phrases ("This skill should be used when the user asks to..."), and the "what" is stated directly (integrating MCP servers into Claude Code plugins for external tool and service integration). Not score 4, because the "when" clause is highly explicit and multi-trigger rather than "could be more specific".

5 / 5

Trigger Term Quality

Trigger coverage is comprehensive and natural: "add MCP server", "integrate MCP", "set up Model Context Protocol", "use .mcp.json", "connect external service", "${CLAUDE_PLUGIN_ROOT} with MCP", plus the server-type terms (SSE, stdio, HTTP, WebSocket). Includes full protocol name, the acronym implied by "MCP server", a concrete file extension, and transport-specific jargon a user would actually say.

5 / 5

Distinctiveness Conflict Risk

The niche is sharply defined — MCP server integration specifically within Claude Code plugins — with distinctive triggers like ".mcp.json" and "${CLAUDE_PLUGIN_ROOT}" that no other plausible skill would claim. Overlap risk with generic config or API-integration skills is minimal.

5 / 5

Total

18

/

20

Passed

Validation

87%

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

Validation — 14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

skill_md_line_count

SKILL.md is long (555 lines); consider splitting into references/ and linking

Warning

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

14

/

16

Passed

Repository
anthropics/claude-plugins-official
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.