codedb_context
Task-shaped composer — pass a NL task, get keywords + symbol defs + ranked files + top snippets in one block; format=json adds typed provenance, and documenthops=1..2 explicitly expands linked Markdown
How to use it
codedb_context is exposed by the Codedb MCP server. Add the server to your MCP client (Claude Desktop, Cursor, Windsurf and others), and the codedb_context tool becomes available to the model automatically. See the full listing for setup details and every tool this server provides.
Install Codedb
npx -y codedeebee mcpOther tools in Codedb (18)
Every call site of a symbol — word index ∩ outline scope, in one round-trip
Changed files since a sequence number
Typed dependency graph: imports by default, or Markdown links with edgetype=documents; document traversal is capped at 2 hops / 64 files
Fuzzy file-name search (typo-tolerant subsequence match against indexed paths — not a content/symbol search)
Match indexed paths against a glob pattern (src//.zig, .md, …)
Most recently modified files
Index a local folder and write codedb.snapshot
List immediate children of a directory — dirs first, then files with language + counts
Symbols in a file: functions, structs, imports, with line numbers
List all locally indexed projects on this machine
Composable pipeline — chain find, search, filter, deps, outline, read, sort, limit in one request
Read file content (line ranges, ifhash skip-unchanged, compact mode)
Trigram-accelerated full-text search (supports regex, scoped results)
Full pre-rendered JSON snapshot of the codebase
Index status (file count, current sequence, scan phase)
Find where a symbol is defined across the codebase
Full file tree with language, line counts, symbol counts
O(1) inverted index word lookup