Use when renaming symbols across files, finding all references to a function/class, moving code between files, or inlining variables — LSP-backed atomic multi-file refactoring
中文简介正在准备中。当前显示原文说明。
实测评分 未实测
查看安装方法安装前可以查看交给 AI 代理的指令内容。
MCP server key: serena — tools: mcp__serena__*
Core principle: Type-aware, 100% recall. Grep finds strings; Serena finds symbols — including aliased imports, dynamic usages, and type-renamed references.
| Question | Use |
|---|---|
"Rename AuthService everywhere" | serena — rename_symbol |
"Find all callers of login()" | serena — find_references |
"Move UserMapper to another file" | serena — move_symbol |
"Inline the MAX_RETRIES constant" | serena — inline_symbol |
| "Find code about rate limiting" | cocoindex-code — semantic meaning |
| "Who calls AuthService in the graph?" | codebase-memory-mcp — architecture overview |
| "Is renaming this safe to ship?" | code-review-graph — blast radius |
mcp__serena__find_symbol(name, type?) — find by name/type (LSP, not grep)
mcp__serena__find_references(symbol) — all usages across the project
mcp__serena__get_symbol_info(symbol) — signature, docstring, defined-in, refs
mcp__serena__rename_symbol(symbol, new_name) — rename everywhere atomically
mcp__serena__move_symbol(symbol, target_file) — move + fix all imports
mcp__serena__replace_symbol_body(symbol, body) — replace impl, keep signature
mcp__serena__inline_symbol(symbol) — inline at all call sites
mcp__serena__get_call_graph(symbol) — call graph from a symbol
Rename across codebase:
1. mcp__serena__find_symbol("OldName") → confirm it's the right symbol
2. mcp__serena__find_references("OldName") → verify scope
3. mcp__serena__rename_symbol("OldName", "NewName") → atomic rename
Move a class to another file:
1. mcp__serena__get_symbol_info("MyClass") → see current location and imports
2. mcp__serena__move_symbol("MyClass", "src/new_module.py")
Find all callers before deleting:
1. mcp__serena__find_references("legacyFunction") → if empty, safe to delete
.serena/project.yml (committed to repo) controls which language servers are active:
languages:
- python
- typescript
- javascript
Language servers are auto-installed by serena on first use (pyright, typescript-language-server).
Low — on-demand per MCP call. No background indexing process. LSP server initializes on first call (~2-3s), then fast.
codebase-memory-mcp (Cypher graph, whole-codebase view)cocoindex-code (find code by meaning when you don't know exact names)code-review-graph (blast radius, safety score before PR)Grep tool (faster for literal strings with no aliasing concern)还没有评价。欢迎分享你的使用感受。
无需登录即可发表。欢迎分享使用感受、提问，或说明在哪些环境下无法运行。
还没有评论。来写第一条吧。
Route a codebase question to the right tool — structural graph, semantic search, or risk analysis. Use when you need to find or understand code without knowing exact file names.
中文简介正在准备中。当前显示原文说明。
Auto-configure CLAUDE.md and claude/ knowledge docs for a new project. Use when setting up Brain Bootstrap in a new repository. Discovers the codebase, fills in templates, and writes project-specific configuration.
中文简介正在准备中。当前显示原文说明。
Use before any new feature, component, or behavioral change — when you need to explore requirements and design before writing code
中文简介正在准备中。当前显示原文说明。
Activate safety guards for the session. Use /careful before critical operations — blocks rm -rf, DROP TABLE, git push --force, kubectl delete namespace, docker system prune via PreToolUse hook.
中文简介正在准备中。当前显示原文说明。
Generate a user-facing changelog from git commits — categorize, filter noise, translate to user language. Runs in isolated fork context.
中文简介正在准备中。当前显示原文说明。
Save session state before context gets full or before ending. Writes current task state, branch, and loaded docs to claude/tasks/todo.md so the next session can resume cleanly.
中文简介正在准备中。当前显示原文说明。