second-brain-ingest — Claude Code skill: raw/ files into wiki pages
This skill turns raw documents in raw/ into linked wiki pages. Useful when you keep clipping articles and notes into one folder.
npx skills add NicholasSpisak/second-brain --skill second-brain-ingest -a claude-codeWhat second-brain-ingest does
The skill walks the files in raw/ and builds a summary page in wiki/sources/ for each one. It reads the whole source first, then shows you the 3-5 main takeaways and waits for your confirmation — nothing gets written before that. The summary page holds a structured overview, key claims, and the entities and concepts the source mentions.
From there it updates the entity and concept pages: people, organizations and products go to wiki/entities/, ideas and frameworks to wiki/concepts/. If a page already exists, the skill appends the new material and adds the source to the sources: field. When fresh information contradicts what is already there, it edits the page and flags the contradiction with both sources cited. Cross-references use [[wikilinks]] rather than file paths.
Finally it updates wiki/index.md and wiki/log.md. The log is what makes repeat runs safe: files already listed there are skipped, so you can drop in a new article and run the skill again. One source usually touches 10-15 wiki pages, which is expected. Note that the skill only works on files you have already placed in raw/ — with no such folder there is nothing to process.
What it helps with
- file a downloaded article into the wiki
- spot files in
raw/that were never ingested - link a new entity to pages where it already appears
- record a contradiction between a new source and an older note
- refresh the index and log after adding material
Example prompts for the agent
- process the new sources in raw/
- ingest this article, I dropped it into raw/
- go through this file and tell me what matters in it
How to install second-brain-ingest
1Skills CLI (Vercel)
npx skills add NicholasSpisak/second-brain --skill second-brain-ingest -a claude-code2Manually: put the files in place
~/.claude/skills/second-brain-ingest/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/second-brain-ingest/
1Skills CLI (Vercel)
npx skills add NicholasSpisak/second-brain --skill second-brain-ingest -a codex2Manually: put the files in place
.agents/skills/second-brain-ingest/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add NicholasSpisak/second-brain --skill second-brain-ingest -a cursor2Manually: put the files in place
.cursor/skills/second-brain-ingest/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install NicholasSpisak/second-brain second-brain-ingest2Manually: put the files in place
.github/skills/second-brain-ingest/SKILL.md1Skills CLI (Vercel)
npx skills add NicholasSpisak/second-brain --skill second-brain-ingest -a gemini-cli2Platform CLI
gemini skills install https://github.com/NicholasSpisak/second-brain --path skills/second-brain-ingestManually: put the files in place
.devin/skills/second-brain-ingest/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add NicholasSpisak/second-brain --skill second-brain-ingest -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about second-brain-ingest
How do I install second-brain-ingest in Claude?
Run npx skills add NicholasSpisak/second-brain --skill second-brain-ingest -a claude-code. The skill lands in ~/.claude/skills/second-brain-ingest/SKILL.md.
What does second-brain-ingest do?
It reads documents from raw/, creates a summary page for each in wiki/sources/, and adds the entities and concepts mentioned to their own pages. It finishes by updating the wiki index and log.
How is it different from just saving a note?
It looks for an existing page first and extends it instead of creating a new one per fact. A new page appears only when the topic stands on its own.
