A secure AI reasoning skill that analyzes email conversations to extract tasks, summarize threads, and detect sentiment via the iGPT Context Engine.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install igpt-email-ask
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install igpt-email-ask using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is iGPT Email Ask?
iGPT Email Ask is a specialized tool designed to provide deep, reasoned analysis of email data without requiring direct filesystem or shell access. By utilizing the iGPT Context Engine, this skill reconstructs full conversation threads across replies, forwards, and CCs to identify who made specific decisions and what action items remain open. As a key component of Openclaw Skills, it offers developers a secure way to integrate email intelligence into their agents, providing per-user isolation and OAuth-scoped data handling.
This skill goes beyond simple keyword searching by understanding intent and context over time. It can transform unstructured email bodies into highly structured JSON data, making it an ideal choice for building automated workflows that require grounded answers from a user's communication history. Whether you are preparing a meeting briefing or identifying risks in a sales deal, this skill delivers precise, context-aware insights.
iGPT Email Ask Use Cases
- Summarize complex, long-running email threads into concise executive summaries.
- Automatically extract action items, owners, and deadlines from project-related emails.
- Analyze the sentiment of customer interactions to detect potential churn or deal risks.
- Generate meeting briefings by synthesizing the last 7 days of relevant correspondence.
- Create structured JSON data from unstructured emails for integration into CRMs or task managers.
How iGPT Email Ask Works
- The user sends a natural language prompt or a structured data request to the skill.
- The skill validates the user's identity and checks for an active OAuth connection to their email datasource.
- A request is sent to the iGPT Context Engine's recall/ask endpoint, specifying the desired quality tier (normal, high, or reasoning).
- The engine reconstructs the relevant conversation timelines and identifies key entities and decisions.
- The skill returns a response in the requested format, such as plain text for summaries or schema-validated JSON for structured data extraction.
iGPT Email Ask Setup
To get started with this skill, first install the official Python package:
pip install igptai
Next, configure your environment with your iGPT API key to enable connectivity with Openclaw Skills:
export IGPT_API_KEY="your-api-key-here"
Users must authorize their email account via OAuth before the skill can retrieve data. You can verify connection status using the datasources.list() method within your agent logic.
iGPT Email Ask Data Schema & Taxonomy
The skill facilitates structured data exchange through flexible output formats and schema validation. Below is the primary parameter structure for queries:
| Parameter | Type | Description |
|---|---|---|
| input | string | The natural language question or analysis prompt. |
| quality | string | Tier choice: cef-1-normal, cef-1-high, or cef-1-reasoning. |
| output_format | string/object | Supports text, json, or a strict JSON Schema definition. |
| stream | boolean | Enables SSE streaming for real-time parsed JSON chunks. |
iGPT Email Ask Advanced Features
- Contextual Reasoning Tiers: Select from different quality levels to balance speed versus depth of analysis for complex cross-thread synthesis.
- Schema-Validated Output: Enforce strict JSON schemas to ensure the AI returns data that perfectly fits your application's data structures.
- SSE Streaming: Supports Server-Sent Events to stream parsed JSON chunks, improving perceived latency in interactive applications.
- Per-User Isolation: Native support for user-scoped indexing ensures that data is never leaked between different users within the same environment.
- Intent Reconstruction: Automatically identifies ownership and decision-making history within messy email threads.
SKILL.md
Loading
Related Openclaw Skills
reMarkable Tablet
Bridge your reMarkable tablet to your AI workspace to browse, search, and OCR handwritten notes seamlessly.
Groq Completion
A high-performance AI inference skill that connects your agent to Groq's ultra-fast LPU-powered API for near-instant completions.
Groq Completion
A high-performance integration that leverages Groq's LPU technology for near-instantaneous AI text generation and completions.
Ironclaw
Ironclaw is a real-time safety and threat classification engine designed to protect AI agents from malicious content and unauthorized data exfiltration.
iGPT Email Search
Secure, per-user semantic email search across Gmail, Outlook, and IMAP histories.
iGPT Email Search
A secure, per-user-isolated semantic search tool for retrieving relevant emails and threads from Gmail, Outlook, or IMAP via the iGPT API.
Featured*
