codetex-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| CODETEX_DATA_DIR | No | Base directory for storage (SQLite DB and cloned repos). Defaults to ~/.codetex. | |
| ANTHROPIC_API_KEY | Yes | Anthropic API key for LLM summarization. Required for indexing and most MCP tools. | |
| CODETEX_LLM_MODEL | No | Model used for summarization (e.g., claude-sonnet-4-5-20250929). | |
| CODETEX_LLM_PROVIDER | No | LLM provider. Currently only 'anthropic'. | |
| CODETEX_EMBEDDING_MODEL | No | Sentence-transformers model for embeddings. Default: all-MiniLM-L6-v2. | |
| CODETEX_TIER1_THRESHOLD | No | Threshold ratio to rebuild repo overview on sync. Default: 0.10. | |
| CODETEX_MAX_FILE_SIZE_KB | No | Maximum file size in KB to index. Default: 512. | |
| CODETEX_MAX_CONCURRENT_LLM | No | Maximum concurrent LLM calls during indexing. Default: 5. |
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 |
|---|---|
| get_repo_overviewA | Return the Tier 1 markdown overview for a repository. Includes repository purpose, directory structure, key technologies, entry points, and architecture patterns. |
| get_file_contextC | Return the Tier 2 file summary for a specific file. Includes file purpose, public interfaces, dependencies, role classification, line count, and token count. |
| get_symbol_detailC | Return the Tier 3 detail for a specific symbol. Includes full signature, description, parameters with types, return type, call relationships, and file location. |
| search_contextC | Search for relevant code context using semantic similarity. Returns a ranked list of matching files and symbols with relevance scores. |
| get_repo_statusB | Return index status for a repository. Includes indexed commit, current HEAD, staleness indicator, file count, symbol count, total tokens, and last indexed time. |
| sync_repoB | Trigger an incremental sync for a repository. Processes only files changed since the last indexed commit. Returns a summary of changes made. |
| list_reposA | List all registered repositories with their status. Returns a markdown table of repository names, remote URLs, indexed commits, and file counts. |
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 7 tools
Each tool targets a distinct concern: repo-level overview, file context, symbol detail, semantic search, status, syncing, and repo listing. There is no meaningful overlap or ambiguity between tool boundaries.
Tool names consistently follow a snake_case verb_noun pattern, mostly using get_ for retrieval actions. Minor variation like search_context and sync_repo still fit the same predictable style.
Seven tools is well-scoped for a code context and indexing server. Each tool earns its place by covering a distinct retrieval or maintenance operation without unnecessary bloat.
The set covers repository overview, file context, symbol detail, semantic search, status, sync, and listing—covering the core workflows well. The main gap is the lack of explicit repo registration or full-sync tools, though these may be handled externally.