feat: KBDB self-hosted 查詢 + embed 模組 + thin-shell 收窄 + search_workflow(code done 待端到端)
按 issue 分段標明(檔 #5/#8 改動交疊處無法乾淨拆檔,故併一個 commit): #4 thin-shell §3.1 自力救濟階梯 + code-node 規則(純文檔/規則,code-node 零件未實作) #5 KBDB source filter(json_extract metadata_json 零建表)+ 能力對照;documents 聚合與 DELETE proxy 部分擱置等頂層 T8 #7 base embed 模組(kbdb/src/embed.ts)+ vectorize 開關(deploy/config/wrangler.toml 註解範本) + 語義查詢降級閉環(mode=semantic 未開→LIKE+capability_hint) #8 部分(workflow-discovery): - KBDB /entries/search 加 base 通用 entry_type filter(entry-crud/embed/route/kbdb-proxy 透傳) - /webhooks/named 強制 description(空→400,訊息要求操盤 AI 據實寫一句) - 部署雙寫 entry_type=workflow embeddable entry(waitUntil 非阻塞,供 search) - cypher GET /workflows/search + MCP u6u_search_workflows(優先語意、降級 hint) - cypher POST /workflows/backfill-search-entries(無 desc 列出不編造) - GET /webhooks/named 補回 description/created_at 欄位(為 list 來源收斂備) ⚠️ tsc 綠 = code done,非完成(mindset §7 禁假綠): - #7/#8 端到端待 leo21c 部署驗(Vectorize 需官方憑證、CC 跑不了) - #8 ①-a(MCP deploy 改打 /webhooks/named)未做、MCP deploy 那半仍 404 - #8 端到端(強制填擋空/語義命中/租戶隔離/降級 hint)未驗 Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This commit is contained in:
@@ -170,23 +170,39 @@ export function registerQuery(server: McpServer, env: Env) {
|
||||
);
|
||||
}
|
||||
|
||||
/** kbdb_search — 對 entries 做 D1 LIKE 關鍵字搜尋(基本盤,非語義)。 */
|
||||
/**
|
||||
* kbdb_search — 對 entries 做搜尋。mode=keyword(D1 LIKE,基本盤永遠可用)或 semantic(向量,需開 embed 模組)。
|
||||
* 語義/關鍵字都在同一 KBDB MCP(用戶資料 RAG),不分散(issue #7 / D17 邊界)。
|
||||
* mode=semantic 但沒開 vectorize → base 自動降級 keyword + 回 capability_hint(發現閉環,叫 CC 幫開)。
|
||||
*/
|
||||
export function registerSearch(server: McpServer, env: Env) {
|
||||
server.tool(
|
||||
"kbdb_search",
|
||||
"對 KBDB 內容做關鍵字搜尋(D1 LIKE,基本盤層;語義搜尋是另外的 embed 模組,基本盤沒有)。" +
|
||||
"回命中的 entries。要按 template 取整批結構化資料用 kbdb_query。",
|
||||
"搜尋 KBDB 內容。mode='keyword'(預設,D1 LIKE 關鍵字,基本盤永遠可用)或 'semantic'(AI 向量語義搜尋," +
|
||||
"需先開 embed 模組)。語義沒開時會自動降級關鍵字並告訴你怎麼開。要按 template 取整批結構化資料用 kbdb_query。",
|
||||
{
|
||||
q: z.string().min(1).describe("搜尋關鍵字"),
|
||||
q: z.string().min(1).describe("搜尋關鍵字 / 語義查詢句"),
|
||||
owner_id: z.string().optional().describe("限定某歸屬範圍內搜(選填)"),
|
||||
source: z.string().optional().describe("只搜某來源(ingest source.uri,選填)"),
|
||||
mode: z.enum(["keyword", "semantic"]).optional().describe("keyword(預設)或 semantic(需開 vectorize)"),
|
||||
},
|
||||
async ({ q, owner_id }) => {
|
||||
async ({ q, owner_id, source, mode }) => {
|
||||
try {
|
||||
const path = `/entries/search?q=${encodeURIComponent(q)}` + (owner_id ? `&owner_id=${encodeURIComponent(owner_id)}` : "");
|
||||
const res = await kbdbFetch(env, path);
|
||||
const qs = new URLSearchParams({ q });
|
||||
if (owner_id) qs.set("owner_id", owner_id);
|
||||
if (source) qs.set("source", source);
|
||||
if (mode) qs.set("mode", mode);
|
||||
const res = await kbdbFetch(env, `/entries/search?${qs.toString()}`);
|
||||
if (!res.ok) return errorResponse("search_failed", `搜尋失敗`, ["稍後重試"], await res.text().catch(() => ""));
|
||||
const data = await res.json();
|
||||
return successResponse(data, ["mode:keyword = D1 LIKE(基本盤)", "找不到時換個關鍵字,或用 kbdb_query 按 template 列出"]);
|
||||
const data = (await res.json()) as { mode?: string; capability_hint?: string };
|
||||
// base 回 capability_hint → 語義沒開、已降級 keyword。把它當 next-step 傳給 AI(發現閉環)。
|
||||
const hints =
|
||||
data.capability_hint
|
||||
? [data.capability_hint, "要開:跟用戶確認後,CC 可代開(寫 config kbdb_embed:true + acr update)"]
|
||||
: data.mode === "semantic"
|
||||
? ["mode:semantic = AI 向量語義搜尋"]
|
||||
: ["mode:keyword = D1 LIKE(基本盤)", "想要語義搜尋:mode='semantic'(需先開 vectorize)"];
|
||||
return successResponse(data, hints);
|
||||
} catch (e) {
|
||||
return errorResponse("internal_error", e instanceof Error ? e.message : String(e), ["稍後重試"]);
|
||||
}
|
||||
|
||||
Reference in New Issue
Block a user