khanote.research.ingest

SkillDocs & knowledge

Your research was almost Khan-worthy. Now it is. You're welcome.

Use khanote.research.ingest in Claude, ChatGPT or Ahel Desktop

Free. Sign in, add khanote.research.ingest and connect your AI. About a minute.

Then ask your AI: use the khanote.research.ingest skill

Details

Instructions available. Your AI can read the instructions. Execution depends on the setup they require.

Account requirements not reviewed. Check the skill instructions before use; Ahel provides instructions and does not run this skill.

No other account needed.

Add Ahel to your AI once: Claude, ChatGPT, Cursor, Claude Code or Codex. Then ask it to use this.

khanote.research.ingestStart free

The instructions your AI receives, as published by almost-khan/khanote in skills/khanote.research.ingest/SKILL.md and read by Ahel’s review.

You are ingesting sources into the active research session.

Read {vault_path}/.khanote/current_session to find the active session path. If it does not exist, ask the user to run /khanote.research.start first.
Collect sources from the user — accept any combination of:
- URLs (web pages, arXiv links, etc.)
- Local file paths (PDF, Markdown, text)
- Inline text pasted directly into the conversation
For each source:
- Copy files to the session's sources/ directory.
- Download URLs to sources/ as .html or .pdf where possible.
- Record the source in _session.md under ## Sources with its type and timestamp.
Send each source to the active researcher for indexing. Use the default researcher from config.yaml unless the user specifies otherwise.
If a researcher is unavailable (no API key or network error), record the source with status pending — do not fail.
Tell the user how many sources were ingested and any that are pending. Suggest running /khanote.research.analyze next.

Advanced