Template for skills that integrate with an MCP server. Demonstrates the reference file pattern: Claude reads a domain-specific MCP cheatsheet before making any tool calls, reducing query failures caused by server-specific gotchas. Fork this skill and replace the Sentry example with your target MCP.
71%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
Low
Low-risk findings worth noting
Fix and improve this skill with Tessl
tessl review fix ./examples/skills/mcp-integration-reference/SKILL.mdThe skill exposes the agent to untrusted, user-generated content from public third-party sources, creating a risk of indirect prompt injection. This includes browsing arbitrary URLs, reading social media posts or forum comments, and analyzing content from unknown websites.
The required workflow bulk-fetches unresolved Sentry issues and full stack traces via `list_issues` and `get_issue` without the agent selecting a specific item, and Sentry error messages and event data can contain input that outsiders control through public-facing applications.
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.