brain-ingest-pipeline

Processes email, calendar, and meeting data into structured brain knowledge through a five-phase pipeline.

1|Updated Jun 21, 2026
One-click install
npx skills add https://github.com/tapway/shogun-os --skill brain-ingest-pipeline-tapway
Or copy as Structured Prompt for Agentβ–Ό
Please help me install this Agent Skill.
Skill: brain-ingest-pipeline
Source: https://github.com/tapway/shogun-os/tree/main/skills/gbrain/brain-ingest-pipeline
Command: npx skills add https://github.com/tapway/shogun-os --skill brain-ingest-pipeline-tapway

SYSTEM DOCUMENTATION & REQUIREMENTS

πŸ’‘ This Skill requires google-api-python-client, google-auth, and includes scripts (resource) components.

What problem does it solve? Ingesting emails, calendar events, and meeting notes into a knowledge graph often results in orphaned pages, missing links, and unvalidated content. This Skill enforces a single COLLECT β†’ ROUTE β†’ BRIDGE β†’ ENRICH β†’ VALIDATE flow so every ingest source produces linked, enriched, and compliance-checked brain pages. ## Core Features & Use Cases - Unified Five-Phase Pipeline: Collects raw data via scripts, routes signals to deals/projects/HR pages, bridges entities with graph links, enriches person profiles, and validates every modified page. - Multi-Account Gmail Triage: Labels and prioritizes inboxes across team accounts using service account domain-wide delegation with batch rotation. - Calendar Collection: Exports events from all team calendars into gbrain-indexable markdown with PII scrubbing. - Use Case: A nightly cron runs gmail-triage across 10 accounts, routes sales signals to matching deal pages, creates person-to-company graph links, flags stalled deals, and outputs a structured pipeline summary. ## Quick Start Ask the agent to run the brain ingest pipeline to process today's emails and calendar events into the brain.

Frequently Asked Questions about brain-ingest-pipeline

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

FAQPage Schema
How do I ingest Gmail emails into a knowledge graph automatically?β–Ό

Run the gmail-triage.py script via cron to label and collect inbox messages, then route signals to matching deal or project pages using gbrain MCP queries. The pipeline creates graph links, adds timeline entries, and validates every modified page.

How to sync Google Calendar events from multiple team accounts?β–Ό

Use collect-calendar.py with a service account configured for domain-wide delegation. It impersonates each team member, pulls events from a 7-day lookback to 14-day lookahead window, and writes PII-scrubbed markdown files to the brain data directory.

What is service account domain-wide delegation for Gmail?β–Ό

Domain-wide delegation lets one Google service account access Gmail and Calendar data for every user in a Google Workspace domain without individual OAuth consent. The scripts impersonate each user via the subject parameter when building API clients.

Why are ingested pages not linked in the brain graph?β–Ό

Missing links usually mean the BRIDGE phase was skipped or entity resolution failed. Use mcp_gbrain_resolve_slugs to find matching person and company pages, then create links with mcp_gbrain_add_link before running validation.

When should I not use this ingest pipeline?β–Ό

Skip it for single manual brain page writes, which should use brain-compliance directly, and for media or file ingestion handled by dedicated media ingest skills. It is designed for batch email, calendar, and meeting processing only.