secondbrain_skill

This skill helps you capture ideas, tasks, and notes in real time, organize them, and surface insights from your knowledge graph.
  • Python

20

GitHub Stars

2

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill leegonzales/aiskills --skill secondbrain

  • README.md1.3 KB
  • SKILL.md7.4 KB

Overview

This skill is a conversational Second Brain for capturing thoughts, managing knowledge, and surfacing actionable insights during work sessions. It captures ideas, tasks, meetings, and references with minimal interruption and keeps a searchable knowledge graph. Use it to triage an inbox, generate daily digests, query your notes semantically, or update task status quickly.

How this skill works

During conversation the skill recognizes capture intent (explicit or implicit), extracts context, classifies the capture into typed nodes (task, idea, reference, meeting, etc.), and either auto-saves or sends low-confidence items to review. It supports semantic search, graph traversal across relationships (supports, blocks, contains), and short digests that surface due items and priorities. Actions let you complete tasks, change priority, archive nodes, and link items to meetings or people.

When to use it

  • When you want to save an idea or decision mid-conversation ("remember this")
  • To add or update a task quickly ("add a task", "mark complete")
  • To check pending captures or triage your inbox ("show my inbox")
  • When you need a focused daily or weekly digest ("what should I focus on today?")
  • To query past notes, decisions, or relationships ("what did I say about authentication?")

Best practices

  • Capture only meaningful items to avoid clutter; be selective when intent is weak
  • Provide minimal context (project, due date, assignee) when creating tasks to improve classification
  • Review low-confidence captures in the inbox to correct types or add links
  • Link new captures to existing nodes to prevent duplicates and preserve graph integrity
  • Use short commands for quick actions (capture, inbox, digest, done <id>)

Example use cases

  • Capture a deadline mentioned in conversation and create a task with a due date
  • Paste a meeting transcript to auto-extract tasks, decisions, people, and insights
  • Ask for today’s digest to see overdue and high-priority items before starting work
  • Search the knowledge graph for all references supporting a specific goal
  • Mark a PR review task complete or change its priority during a session

FAQ

Items below the confidence threshold are routed to the inbox with a needs_review state so you can confirm or reclassify them.

Can meeting transcripts create tasks automatically?

Yes — transcripts are parsed into meeting summaries, task action items, decisions as references, person links, and insights, then presented for confirmation.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational