Skip to content
TakoAPI
Back to skills

igpt-email-ask

by sammy-spk

Search & Research
355Downloads reported by clawskills.sh
2Stars
0Views
0 likesSign in to like this skill
—Not yet rated
Your rating

Brief

Secure, per-user-isolated email reasoning and analysis via the iGPT Context Engine API.

Description

What This Skill Does

Queries iGPT's Context Engine API to analyze a user's connected email data. Reconstructs full conversation threads to identify decisions, ownership, tasks, and sentiment across replies, forwards, and CCs. Returns plain text, JSON, or schema-validated structured output with optional streaming.

Unlike keyword search or basic retrieval, the Context Engine reconstructs full thread history across replies, forwards, and CCs to reason about ownership, decisions, and risk across time.

When to Use It

  • Extracting open action items from this week's email threads
  • Summarizing deal correspondence before a sales call
  • Finding who owns a pending approval and when it's due
  • Detecting negative tone shifts in customer threads over the past week
  • Generating a pre-meeting briefing from recent email exchanges

Example Workflow

INPUT

User asks: What are the open action items from this week's board meeting?

AGENT
  1. 1

    Authenticates to the iGPT API using IGPT_API_KEY

  2. 2

    Queries the recall/ask endpoint with the user's question and a schema-validated output format

  3. 3

    Context Engine reconstructs relevant email threads and identifies task-related content

  4. 4

    Returns a JSON object with action items, owners, and due dates

OUTPUT

JSON with an action_items array, each entry containing title, owner, and due_date fields

Requirements

  • IGPT_API_KEY environment variable (obtain at igpt.ai/hub/apikeys/)
  • iGPT account with email datasource connected via OAuth authorization

Installation

clawhub install sammy-spk/igpt-email-ask

Run the CLI command, or paste the prompt into Claude Code, Cursor, Codex or whichever coding agent you use.

Links