Files
ask262/.opencode/tools/ask262-search-spec-sections.ts
T

44 lines
1.4 KiB
TypeScript

/**
* OpenCode custom tool: ask262_search_spec_sections
* Searches the ECMAScript specification for relevant sections.
*/
import { tool } from "@opencode-ai/plugin";
import * as lancedbSdk from "@lancedb/lancedb";
import { OllamaEmbeddings } from "@langchain/ollama";
import { createSearchSpecSectionsTool } from "../../src/agent-tools";
export default tool({
description:
"Vector search the ECMAScript specification for sections relevant to a query. " +
"Returns JSON array with sectionId, sectionTitle, score, partIndex, totalParts, and content. " +
"Use this when you need to find spec sections related to a JavaScript topic or question.",
args: {
query: tool.schema
.string()
.describe(
"The search query to find relevant specification sections (e.g., 'how does array map work')",
),
},
async execute(args, context) {
const { query } = args;
const { worktree } = context;
// Initialize embeddings
const embeddings = new OllamaEmbeddings({
model: "qwen3-embedding:0.6b",
});
// Connect to LanceDB
const storageDir = `${worktree}/storage`;
const db = await lancedbSdk.connect(storageDir);
const table = await db.openTable("spec_vectors");
// Create and execute tool
const searchTool = createSearchSpecSectionsTool(table, embeddings);
const result = await searchTool.func({ query });
return result;
},
});