AI & Machine Learning

so-yesterday.ai

Provides access to an AI knowledge base with daily digests, videos, essays, and knowledge concepts for AI learning.

1 endpoint13 known toolsFirst detected May 11, 2026Last detected August 30, 2026

ENDPOINT 1

https://so-yesterday.ai/mcp

No auth detected

MCP server metadata

Name: so-yesterday
Version: 1.29.0

Capabilities

experimentalpromptsresourcestools

Server instructions

This MCP server provides access to an AI transformation knowledge base. Read tools are anonymous. Write tools require a Bearer soat_… token issued via the OAuth consent flow at https://so-yesterday.ai. Start with the 'search' tool or 'get_latest_digest' for recent developments.

Known tools 13

get_latest

Get a one-call briefing: latest daily digest (full content), 5 recent highlighted videos with summaries, 3 latest essays, and knowledge base stats.

Inferred read-only

search

Search the AI knowledge base across all content types (videos, knowledge concepts, essays, digests, personas).

Inferred read-only

get_video

Get a video's executive summary, key points, and optionally the full transcript.

Inferred read-only

list_videos

Browse videos with optional filters by channel or tag.

Inferred read-only

get_knowledge

Read a knowledge concept article with related concept links.

Inferred read-only

get_essay

Read a full original essay on AI transformation.

Inferred read-only

get_digest

Read a daily digest of the 5 most important AI developments for a specific date.

Inferred read-only

get_latest_digest

Get the most recent daily digest.

Inferred read-only

list_tags

List all content tags with usage counts.

Inferred read-only

get_knowledge_graph

Get the knowledge graph data — nodes (concepts) and edges (relationships).

Inferred read-only

put_personal_note

Create or update a personal knowledge note.

Potential side effects

create_post

Create and immediately submit a post for moderation review.

Potential side effects

submit_post_tool

Submit an existing draft post for moderation review.

Potential side effects

CONNECT WITH APPROVAL

Client installation

Review this server and its permissions before adding it. Secret placeholders must be set locally.

Codex

~/.codex/config.toml

[mcp_servers.so-yesterday]
url = "https://so-yesterday.ai/mcp"
enabled = true

Claude Code

.mcp.json

{
  "mcpServers": {
    "so-yesterday": {
      "type": "http",
      "url": "https://so-yesterday.ai/mcp"
    }
  }
}

Claude Desktop

Settings → Connectors → Add custom connector

Name: so-yesterday
Remote MCP URL: https://so-yesterday.ai/mcp

Add this remote URL as a custom connector in Claude Desktop. Availability depends on the user plan and workspace policy.

Cursor

.cursor/mcp.json

{
  "mcpServers": {
    "so-yesterday": {
      "url": "https://so-yesterday.ai/mcp"
    }
  }
}

Visual Studio Code

.vscode/mcp.json

Add to Visual Studio Code

{
  "servers": {
    "so-yesterday": {
      "type": "http",
      "url": "https://so-yesterday.ai/mcp"
    }
  }
}

Generic MCP

Client-specific MCP configuration

{
  "name": "so-yesterday",
  "transport": "streamable-http",
  "url": "https://so-yesterday.ai/mcp"
}

MCP Inspector

Run the official MCP Inspector locally and enter the indexed Streamable HTTP endpoint.

TRUST AND VERIFICATION EVIDENCE

Loading Trust v2 evidence…

Checking the associated registrable domain. The BuiltWith key remains server-side.

Indexed

Evidence is source-attributed and does not guarantee that a third-party server is safe. Risk labels are conservative metadata heuristics.