token-context-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| TOKEN_CONTEXT_CONFIG | No | Optional path to an explicit shared/portable TOML configuration file for the repository registry. By default, the registry is stored globally for the current user at %APPDATA%\token-context-mcp\repos.toml. |
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": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| list_repositoriesA | List registered repository IDs only. Call this first; roots are never exposed. |
| get_repo_mapA | Return ranked definitions for a repo_id from list_repositories within a bounded context budget. Compact entries are [short_id, path:line, kind/name, optional rank marker]; request format='full' for signatures, per-symbol evidence, and detailed rank_basis. Use for orientation, not proof of full coverage. |
| find_symbolsC | Find source-backed symbols by name or qualified-name fragment. Returns IDs and spans, never arbitrary files. |
| get_module_dependentsB | Return Tree-sitter-extracted lexical import relationships for one indexed path or module. This is not semantic import resolution or lexical call-graph inference; dynamic imports are flagged rather than resolved. |
| search_sourceA | Search indexed symbol bodies with FTS5 and return bounded source snippets, symbol IDs and line evidence. |
| get_file_skeletonA | Return imports and source-backed headers from one indexed repository-relative file. Function bodies are elided by default. |
| get_symbol_contextB | Return a bounded source packet around one indexed symbol and observed graph edges. Use original source when body, freshness or ambiguity requires it. |
| get_impact_sliceB | Traverse observed caller/callee edges from a symbol. It is a candidate impact slice, never a proof of complete blast radius. |
| get_index_statusA | Return active snapshot metadata and paths changed since indexing. Run before relying on graph results. |
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 9 tools
Each tool targets a distinct capability—repo enumeration, context maps, symbol lookup, impact slices, index freshness, imports, FTS search, file skeletons, and symbol packets—though find_symbols/search_source and get_impact_slice/get_module_dependents sit close enough that an agent may need careful descriptions. Overall boundaries are clear and the descriptions reinforce purpose.
The set mostly follows a get_<object> pattern with list_repositories, find_symbols, and search_source as reasonable verb variations. All names are snake_case and consistently place the action before the object, creating a predictable surface.
Nine tools is appropriate for a token-context indexing server: each tool covers a distinct aspect of repository context without redundancy. The count feels neither thin nor overloaded.
The surface covers the full workflow: list available repositories, fetch orientation maps, search for symbols and source text, inspect imports and file skeletons, check index freshness, and retrieve bounded context packets. A raw full-file read tool is intentionally absent given the bounded-context purpose, but this is a reasonable design choice rather than a gap.