get_v1_code_search
Finds code in the caller org's index across three orthogonal retrieval tiers fused by reciprocal-rank fusion: lexical (FTS5 trigram over code-tokenized text), symbolic (real definition and reference…
Finds code in the caller org's index across three orthogonal retrieval
tiers fused by reciprocal-rank fusion: lexical (FTS5 trigram over
code-tokenized text), symbolic (real definition and reference edges), and
semantic (embedding cosine over AST-boundary chunks). Pick one tier with
type, or leave it to run all three as hybrid, which is what a coding agent
usually wants. It is FAIL-HONEST: a retrieval outage answers 200 with an empty
result set and "degraded": true rather than a 5xx, so an agent degrades instead
of stalling. A malformed regex is a 400.
| Tool | get_v1_code_search |
| Door | https://api.hanzo.ai/v1/mcp |
| Method | tools/call (JSON-RPC 2.0) |
| Arguments | 4 |
| Operation | GET /v1/code/search |
| Product | code |
Arguments
| Field | Type | Required | Default | Values | Description |
|---|---|---|---|---|---|
limit | integer | — | — | — | Limit caps how many spans come back: default 20, maximum 100. A value that is not a positive integer reads as the default. |
q | string | — | — | — | Q is the search query. Required, max 4000 bytes. For type=regex it is a regular expression; for type=symbol it is a symbol name. |
repo | string | — | — | — | Repo narrows to one repository. Empty searches every repo the org has indexed. |
type | string | — | — | — | Type selects the retrieval tier: "text" (FTS5 trigram), "regex", "symbol" (definitions), "semantic" (embeddings) or "hybrid". Anything else — including empty — reads as hybrid. |
tools/list declares a type and a description for each field and nothing further. A — means neither the door nor that operation constrains the field.
Call it
A tools/call carries every argument in one flat object — nothing binds to a path or a query string. Nothing above is required, so every declared argument is shown rather than a guess at which matter.
curl -X POST https://api.hanzo.ai/v1/mcp \
-H "Authorization: Bearer $HANZO_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"jsonrpc": "2.0",
"id": 1,
"method": "tools/call",
"params": {
"name": "get_v1_code_search",
"arguments": {
"limit": 0,
"q": "<q>",
"repo": "<repo>",
"type": "<type>"
}
}
}'Values are the operation's own defaults and enumerated values where it declares them, and a <placeholder> where neither source declares one. limit holds a stand-in that cannot be spelled that way — JSON gives a number, a boolean and a timestamp no placeholder form — so that value is this page's, not the API's. Neither the door nor the operation declares one. tools/list needs no credential; tools/call does — called without one the door answers HTTP 200 with a JSON-RPC result whose isError is set and whose text says what was missing. How to get a key →
The operation behind it
| Operation | Route | Product | Summary |
|---|---|---|---|
get_v1_code_search | GET /v1/code/search | code | Finds code in the caller org's index across three orthogonal retrieval tiers fused by… |
The same capability over plain HTTP is in the code API reference, on https://api.hanzo.ai.
All 755 tools · The door · API reference
Generated from tools/list on https://api.hanzo.ai/v1/mcp — 833 tools captured 2026-08-01, of which 755 are documented here (the operator surface is not published) (this build read the vendored copy; the door was unreachable).
How is this guide?
get_v1_code_file
Returns the INDEXED content of one file — read_file over the chunks the search tiers hold, for pulling up code an agent just found.
get_v1_code_tree
Returns one repository's file structure with a per-file symbol count — get_repo_structure over the org's own index, with no git checkout involved.