Could you add support of *.vue files? To be able to use Serena with enterprise projects with Frontend based on VueJS

[it-sha]

Is your feature request related to a problem? Please describe.
Currently Sirena MCP does not index Vue.js single-file components (*.vue). As a result, the TypeScript and JavaScript code inside these files (in <script> / <script setup> blocks) is not indexed.

For enterprise projects where the frontend is implemented using Vue.js SFCs, this means a significant portion of the codebase is effectively invisible to Sirena MCP. In practice, this makes it impossible (or severely limits the ability) to use Sirena MCP and LLM agents on such projects, because they cannot see or reason about a large part of the frontend logic.

---

Describe the solution you'd like
I would like Sirena MCP to support indexing Vue.js single-file components (*.vue) out of the box.

In particular:

1. Include .vue files in the indexable set

   • Add *.vue to the list of files processed by the indexer (possibly behind a configuration option if needed).

2. Parse and index script blocks

   • Parse .vue files and extract:
     • <script>
     • <script setup>
   • Respect the lang attribute:
     • lang="ts" → process as TypeScript
     • lang="js" (or no lang) → process as JavaScript
   • Feed the extracted code into the existing TS/JS indexing pipeline, so that symbols, references, and types from .vue files are available in the same index as regular .ts / .js files.

3. Integrate with the existing code graph

   • Ensure that imports/exports and type references from .vue files are properly linked to other files in the project (e.g. shared TypeScript types, API clients, stores, etc.).
   • This will let LLM agents traverse the graph across backend and frontend code and solve tasks that span both sides of the stack.

[opcode81]

There is already a PR on this: #784

[opcode81]

Resolved by #784