--- title: "What Medusa Teaches Every MCP Author About Feedback" url: https://daily.dev/posts/what-medusa-teaches-every-mcp-author-about-feedback-gzpfa9ges source_url: https://apievangelist.com/2026/08/26/what-medusa-teaches-mcp-authors-about-feedback type: article source: "API Evangelist" published: 2026-08-26T10:39:47.239Z updated: 2026-08-26T18:30:37.179Z tags: ["architecture", "ai-agents", "mcp"] reading_time: 6 upvotes: 0 comments: 0 language: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # What Medusa Teaches Every MCP Author About Feedback **[API Evangelist](https://daily.dev/sources/apievangelist)** · 6 min read · 0 upvotes · 0 comments ## Summary Medusa, the open-source headless commerce platform, added two features to its remote MCP documentation server: a feedback tool (submit_medusa_feedback) that lets coding agents report broken docs or confusing error flows directly into PostHog and Linear, and an optional context field on tool calls where agents explain why they're making a request. Together these turn the MCP server from a one-way documentation channel into a listening post that surfaces real friction points agents hit while building, such as an incorrectly documented regex matcher in defineMiddlewares or a confusing 401 error on a valid Secret API Key. The piece argues this pattern generalizes: any MCP server can capture agent frustration as structured product signal instead of losing it, since agents report friction agents encounter far more consistently than human developers ever bothered to. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What is the submit_medusa_feedback tool in Medusa's MCP server? It's a dedicated feedback tool Medusa added to its remote MCP documentation server, letting coding agents report concrete problems instead of silently working around them. Reports route directly into PostHog and Linear for triage. It caught a real bug: incorrectly documented defineMiddlewares regex matchers that meant the middleware would never execute. _Developers building MCP servers can track patterns like this on daily.dev to shape their own feedback tooling._ ### Why did Medusa add a context property to its MCP tool calls? The context property lets an agent explain why it's making a request, such as what error a user hit or what concept caused friction, rather than just what it searched for. A bare query like "Secret API key authentication" is ambiguous, but paired with context describing a 401 Unauthorized error on a valid key, it becomes an actionable bug report and prioritization signal. _Anyone designing agent-facing APIs can follow daily.dev for practical patterns like structured context fields._ ## Similar posts on daily.dev - [Seeing MCP](https://daily.dev/posts/seeing-mcp-vfi5ace0s) · API Evangelist · 16 upvotes · 0 comments --- Tags: [#architecture](https://daily.dev/tags/architecture), [#ai-agents](https://daily.dev/tags/ai-agents), [#mcp](https://daily.dev/tags/mcp) [View this post on daily.dev](https://daily.dev/posts/what-medusa-teaches-every-mcp-author-about-feedback-gzpfa9ges)