Wendy - Codebase Q&A Agent
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| ANTHROPIC_API_KEY | Yes | Anthropic API key used to authenticate requests. Loaded from the .env file by python-dotenv. |
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_filesB | List files in a directory, skipping hidden and junk directories. |
| read_fileB | Read the contents of a file, truncated if its too large. |
| searchC | Search all files for a string, returning matching files paths and line numbers. |
| repo_mapC | Show a tree-style map of the project's directories and files. |
| git_logB | Show the most recent git commits, optionally limited to one file. |
| git_blameB | Show who last changed each line of a file, with commit hashes. |
| git_diffA | Show uncommitted changes, optionally limited to one file. |
| list_symbolsC | List the functions and classes defined in a file, with line numbers. |
| find_definitionC | Find where a function or class is defined across the project. |
| semantic_searchC | Find functions/classes semantically related to a natural-language query. |
| verify_claimB | Check whether a natural-language claim about the code is supported or contradicted. Returns a summary plus per-check findings with evidence line numbers. |
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 11 tools
Most tools target clearly distinct operations (file listing, reading, git history, symbol lookup, semantic query, claim verification). The main overlap is between search (literal), semantic_search (semantic), list_symbols (per-file symbols), and find_definition (cross-project definition lookup), but the descriptions draw the boundaries well enough that misselection is unlikely.
All names use snake_case consistently. Most follow a verb_noun pattern (list_files, read_file, find_definition, verify_claim), with a few noun-style names (repo_map, git_log, git_blame, git_diff) and one bare verb (search) that slightly break the pattern but remain readable and predictable.
Eleven tools is well-scoped for a codebase Q&A agent, with each tool covering a distinct retrieval capability (files, content, search, structure, git history, symbols, semantics, verification). No redundant or filler tools are apparent.
The surface covers discovery (list_files, repo_map), reading (read_file), searching (search, semantic_search), structure (list_symbols, find_definition), and history (git_log, git_blame, git_diff), plus claim verification. Minor gaps exist—no regex/glob filtering or call-graph/reference-finding—but core Q&A workflows are fully supported.