Sift — Research intelligence CLI + Tana Bridge

Search article corpora and generate cited research answers via terminal.

1|Updated Mar 21, 2026
One-click install
npx skills add https://github.com/jcfischer/sift-cli --skill sift-research-intelligence-cli-tana-bridge
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: Sift — Research intelligence CLI + Tana Bridge
Source: https://github.com/jcfischer/sift-cli/tree/main
Command: npx skills add https://github.com/jcfischer/sift-cli --skill sift-research-intelligence-cli-tana-bridge

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill solves the problem of finding high-quality, up-to-date information and turning it into actionable answers without manually searching scattered sources or rebuilding context from scratch.

Core Features & Use Cases

  • Search a curated article corpus from the terminal: quickly locate relevant articles by query, time window, and Sift topic.
  • Ask research questions with citations: generate synthesized answers using the article corpus in quick or deep modes.
  • Manage feeds and topics: add sources, assign feeds to topics, and organize ongoing research.
  • Optionally sync results into Tana: import new articles into your Tana nodes as structured children and keep “Last run” up to date.

Quick Start

Ask your AI agent to run Sift for your topic by requesting: "Search Sift for articles about AI security from the last 7 days and return JSON results with titles, URLs, and relevance scores."

Frequently Asked Questions about Sift — Research intelligence CLI + Tana Bridge

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

FAQPage Schema
How do I get cited research answers from the terminal using a CLI?

To get cited research answers, use a terminal-first CLI to query a curated article corpus with your research questions. The system synthesizes corpus-backed responses with citations in quick or deep modes, outputting structured JSON for agent consumption.

Can I sync my research results and imported articles directly into Tana nodes?

Yes, you can sync research results into Tana nodes. The system imports new articles as structured children and updates tracking fields, automatically mapping imported items into user-discovered Tana tags and fields for ongoing research management.

Do I need OAuth credentials to search a research corpus with a CLI?

Yes, you need a local CLI binary and Sift OAuth credentials to search the research corpus. These credentials authenticate your terminal-based requests for topic-scoped keyword or hybrid search and corpus-backed Q&A.

How do I filter article search results by time window and topic?

You can filter article search results by applying time-based filtering and specifying a Sift topic during your CLI query. This narrows the corpus search to recent, relevant articles within your defined research scope.

What is the best way to manage ongoing research feeds and topics from the terminal?

The best way to manage ongoing research feeds is by using the CLI to add sources, assign feeds to specific topics, and organize your research scope. This keeps your article corpus curated and structured for future queries.