code-context-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| database_path | Yes | Path to the SQLite database file where code context will be indexed and stored. |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| find_symbolA | Where is symbol X defined? Lists the files that export a function, class, type or constant by that name. |
| get_file_contextC | What does this file export (signatures, lines), what does it import, and who imports it? |
| who_importsA | Which files import X (file path or exported symbol)? Complete list from the import graph, with symbols, lines and re-exports. No grep needed. |
| exports_inB | What does directory D export, and which exports are unused (unused: true)? Signatures and lines, complete from the import graph. |
| search_filesA | Which files have this text in their path or summary? Not file contents: grep for code. Max 25. |
| queryA | Read-only SQL on the code index: SELECT or PRAGMA table_info(t). files(id, path, language, extension, size_bytes, line_count, summary, description, external_imports, import_lines, created_at, modified_at, indexed_at); exports(id, file_id, name, kind, description, signature, line); dependencies(id, source_id, target_id, symbols): source imports target. *_id = files.id; paths are absolute. Who imports X: dependencies d JOIN files t ON t.id=d.target_id JOIN files s ON s.id=d.source_id WHERE t.path LIKE '%X'. Exports of X: exports e JOIN files f ON f.id=e.file_id. Max 50 rows, 8 KB. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 6 tools
Each tool targets a distinct query shape: symbol definition (find_symbol), file-level context (get_file_context), importers (who_imports), directory exports (exports_in), path/summary search (search_files), and a raw SQL escape hatch. Minor overlap exists because get_file_context already reports importers for one file while who_imports generalizes to symbols, and query can replicate any other tool's answer. Descriptions make the boundaries clear enough that misselection is unlikely.
Names mix styles: verb_noun (find_symbol, get_file_context, search_files), a question-phrase (who_imports), a prepositional form (exports_in), and a bare generic noun (query). Each name is individually readable and reflects its description, but there is no single predictable convention an agent can rely on.
Six tools is a well-scoped set for a code-index server, with no filler entries. Each tool covers a distinct navigation need and the SQL tool absorbs exotic queries instead of spawning extra bespoke tools.
Coverage of the core index domain (locate definitions, describe files, trace imports, list directory exports, search paths, raw SQL) is solid, and query fills most gaps. What's missing is reverse usage lookup beyond imports (call sites/references per symbol) and any way to read actual file contents, though the SQL surface partially compensates.