Reflex
OfficialServer Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
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 | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| search_codeA | Search code for a pattern: every match with path, line and preview (for path and line only, list_locations is cheaper). Whole identifiers by default; contains:true for substrings; regex:true for a Rust regex (alternation, classes, anchors; in JSON double each backslash); symbols:true for definitions only (kind narrows them). Answer: {columns, rows} (each row aligns with columns) plus pagination; when has_more, fetch the next page with offset. total_count is exact only when total_is_exact. |
| list_locationsA | Where does X occur? Every matching line, by file: {path: [line, ...]}; a line with two matches appears once, like grep -n. The cheapest search, no limit. preview:true gives [line, text] pairs (text trimmed, 120 chars). Same matching as search_code. |
| find_referencesA | Every line where a name (any identifier, not only a symbol) is used in code, plus its definition, in code files only. Lines where it is only in a comment or string are left out by a whole-file scan, so no re-check is needed (include_strings:true keeps them). Answer: {definition, references, total_references, returned_count, filtered_out, pagination}; mode:"count" returns the count after filtering. |
| get_dependenciesA | The imports of a file: path, line, internal/external/stdlib. reverse:true lists the files that import it instead; depth:N follows imports N levels (a list of {path, depth}). Static imports only; path may be a fragment or file name. |
| analyzeA | Import-graph analysis. kind: summary (counts), hotspots (most-imported files), circular (import cycles), unused (files nothing imports; entry points included), islands (disconnected groups). Answers are paginated with limit/offset. |
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 5 tools
The tools are mostly distinct, but list_locations and search_code overlap on pattern lookup, and find_references overlaps with search_code's symbol mode. The descriptions clearly distinguish them by cost, output shape, and comment/string filtering, so misselection risk is low but not zero.
All names use snake_case and most follow a clear verb_noun pattern: list_locations, get_dependencies, search_code, find_references. The bare verb analyze is a minor deviation but remains readable and consistent with the set.
Five tools is well-scoped for a code search and import-analysis server. Each tool has a clear role: cheap location lookup, rich search, reference finding, dependency inspection, and graph analysis.
The surface covers code search, occurrence listing, reference lookup, direct imports, and higher-level import graph analysis, which is strong for the apparent domain. Minor gaps remain, such as no direct file-content or project-tree retrieval, though search previews and dependency tools may let agents work around this.