deep-context

Search connected MCP tools in parallel and synthesize cited chronological briefings.

12|4|Updated Mar 14, 2026
One-click install
npx skills add https://github.com/mshadmanrahman/pm-pilot --skill deep-context-mshadmanrahman
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: deep-context
Source: https://github.com/mshadmanrahman/pm-pilot/tree/main/skills/pm-core/deep-context
Command: npx skills add https://github.com/mshadmanrahman/pm-pilot --skill deep-context-mshadmanrahman

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Teams and product managers struggle to assemble a reliable, chronological picture of a topic that spans many siloed tools, which makes onboarding, incident review, and decision-making slow and error-prone. Deep Context automates exhaustive searches across connected MCP systems and synthesizes a single, cited briefing so you can understand history, status, and next steps quickly.

Core Features & Use Cases

  • Parallel cross-tool search: Fan-out searches across Jira, Confluence, Slack, GitHub, Gmail, and Calendar to gather mentions, docs, issues, and communications.
  • Chronological timeline and synthesis: Build an ordered timeline, identify origin, milestones, current state, and surface contradictions with source citations.
  • Stakeholder extraction and artifacts: Identify owners, recent activity, and key documents with links for follow-up.
  • Fallback handling: Detect unreachable internal sources and supplement with web search for public topics, marking any gaps.
  • Use Case: Prepare a briefing for a new PM joining a feature area, investigate a production incident across repos and tickets, or summarize stakeholder decisions before a leadership review.

Quick Start

Run deep-context on "payment retry flow" for the past 90 days with deep depth to produce a timeline, key stakeholders, and source-linked briefing.

Frequently Asked Questions about deep-context

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I synthesize a research briefing across Slack, Jira, and Confluence?

To synthesize a cross-channel research briefing, you can automatically fan out parallel searches across connected MCP tools like Slack, Jira, and Confluence. The system extracts mentions, docs, and communications, assembling them into a single cited timeline.

What is the best way to build a chronological timeline from cross-channel product investigations?

Building a chronological timeline from cross-channel product investigations requires searching connected sources like GitHub and Gmail in parallel. The process identifies milestones, origin points, and current state, while surfacing contradictions with direct source citations.

Can I extract stakeholders and key documents when researching a production incident?

Yes, you can extract stakeholders and key documents when researching a production incident. The system identifies owners, tracks recent activity, and generates links to key artifacts for follow-up alongside the structured briefing.

Does cross-channel research work if internal tools like Jira or Confluence are unreachable?

Cross-channel research includes fallback handling when internal tools like Jira or Confluence are unreachable. It detects unavailable sources, supplements the investigation with web search for public topics, and explicitly marks any information gaps.

How do I prepare a feature area briefing for a new product manager?

Preparing a feature area briefing for a new product manager involves running exhaustive searches across connected MCP systems. It synthesizes history, status, and next steps into a reliable, chronological picture, reducing onboarding time and error-prone manual checks.