[REQUEST] Arcmira: YouTube Transcript Search integration

オープン
  • 何か問題がありますか？
#8,607 コメント 0 件 リアクション 0 件 担当者 0 名 GitHub で見る

メンテナーはふだん 1 日以内に返信

まだ誰も着手していません。

評価

難易度
5/5
見積もり時間
1週間以上
初心者へのやさしさ
35/100
issue の種類
機能追加
明瞭さ
おおむね明確
活発さ
活発
技術スタック
typescript
領域
backend-api-design

調査の方向性

Read the contribution guide for staging and inspect the tool, block, icon, and generated catalog registration requirements first. Use the linked OpenAPI contract and search, authentication, and usage documentation to define the two actions, credential handling, outputs, and error behavior. Done means maintainers agree on the native two-action scope and the required registrations are updated.

索引モデルが issue の本文から書いたものです。

説明

Is your feature request related to a problem? Please describe.

This is a first-party proposal from Arcmira. Sim users can connect Arcmira through the generic MCP or API blocks, but there is no native Arcmira block in the current staging registry. A native block could make timestamped YouTube transcript research available through named actions and documented outputs, without hand-authoring HTTP requests or MCP programs.

Describe the solution you'd like

Add an Arcmira integration, displayed as Arcmira: YouTube Transcript Search where the catalog permits descriptive titles, with two initial actions:

  • Resolve a person, channel or company name through GET /v1/entities/resolve. Return the resolution options instead of silently choosing an ambiguous name.
  • Search spoken transcript passages through GET /v1/search, with a user-controlled result limit and optional documented filters. Preserve timestamped source links, speaker information when available, coverage, partial results and access restrictions.

Use the user's account API key as a user-only configuration input, never a model-supplied argument. Pass credentials in the bearer header. Errors should preserve the API status/code and documentation link, including quota, access and rate-limit failures. The initial block covers entity resolution and indexed search. If expanded to Premium transcript retrieval, paid reads use credits from your plan, then your on-demand budget.

API docs: https://arcmira.com/docs

OpenAPI contract: https://api.arcmira.com/v1/openapi.json

Search guide: https://arcmira.com/docs/search

Authentication and usage: https://arcmira.com/docs/authentication and https://arcmira.com/docs/usage-and-billing

Describe alternatives you've considered

  • Existing remote MCP connection: https://mcp.arcmira.com/mcp. It provides broader research and account operations, with its own tool-effect and entitlement rules: https://arcmira.com/docs/mcp-server
  • Generic API block against https://api.arcmira.com/v1 with the user's bearer key.
  • Separate keyless API-documentation MCP at https://arcmira.com/docs/mcp for developers building integrations. This searches documentation, not video transcripts.

Additional context

Would maintainers prefer a native two-action block or a maintained MCP workflow example? The contribution guide targets staging and requires the tool, block, icon and generated catalog registrations. This proposal seeks agreement on the contribution scope. The local search-and-entity-resolution candidate passed the full repository test suite, repository audits and all-workspace typechecking on October 16, 2026. Both actions also succeeded through the native HTTP routes using an isolated read-only test account, with unchanged account usage. These are dated local results; the candidate still needs a refresh against current staging before contribution. The candidate has not been submitted as a PR or accepted by Sim. Free account API access is available within account allowances, with plan-dependent features and freshness. No credentials or private data are included.

主要言語
TypeScript
スター
29.8k
フォーク
3.9k
平均マージ
1時間 44分
マージ済み PR（30日）
841

環境構築

このプロジェクトには開発コンテナ、Dockerfile、コントリビューションガイドがありません。まず README を読み、一般的な手順ははじめてのコントリビューションガイドを参照してください。

はじめの一歩

  1. issue を最後まで読み、次にプロジェクトのコントリビューションガイドを読みます。
  2. 着手することを issue にコメントします — 二人が同じ作業をするのを防げます。
  3. リポジトリをフォークし、ブランチを切って変更します。
  4. issue 番号を参照したプルリクエストを送ります。

simstudioai/sim のほかの issue

simstudioai/sim の issue をすべて見る

似ている issue

TypeScript の issue をもっと見る

新しい issue をメールで受け取る

初心者向けの GitHub issue を短くまとめたダイジェスト。