knowledge_delta

Compare current findings against a prior state to produce a JSON delta.

1|Updated Mar 23, 2026
One-click install
npx skills add https://github.com/hellonish/singularity --skill knowledge-delta
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: knowledge_delta
Source: https://github.com/hellonish/singularity/tree/main/SKILLS/tier3_output/knowledge_delta
Command: npx skills add https://github.com/hellonish/singularity --skill knowledge-delta

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill identifies and categorizes changes between a newly generated knowledge state and a stored prior baseline, enabling traceable evolution of insights.

Core Features & Use Cases

  • Change detection: Compare two knowledge documents to extract new developments, changed positions, and deprecated claims.
  • Structured outputs: Generate a JSON delta suitable for downstream reporting, dashboards, or knowledge-graph updates.
  • Use cases: Ideal for periodic knowledge updates, competitive intelligence, and regulatory or policy audits where baselines exist.

Quick Start

Provide two inputs—the current findings document and the prior-state artifact identified by prior_date—and run this skill to produce a structured delta in JSON.

Frequently Asked Questions about knowledge_delta

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

FAQPage Schema
How do I compare two JSON knowledge states to find factual differences?

To compare two JSON knowledge states, you provide the current findings document and a prior baseline artifact. The tool identifies semantic and factual differences, outputting a structured JSON delta that tracks new developments and deprecated claims.

What is a knowledge delta and when do I need version control for knowledge management?

A knowledge delta is a structured record of changes between a current knowledge state and a stored baseline. You need it for periodic knowledge updates, competitive intelligence, or compliance audits to ensure traceable evolution of insights.

How do I track new developments and changed positions in dynamic domains?

You track new developments by comparing your current findings document against a stored prior state. This process extracts new developments, changed positions, and deprecated claims, generating a structured JSON delta for monitoring dynamic domains.

Can I generate a structured data diff for compliance audits using a baseline?

Yes, you can generate a structured data diff for compliance audits by comparing current findings against a prior baseline. It requires a plan metadata field named prior_date to identify the prior state artifact for the comparison.

What is the best way to prepare inputs for knowledge comparison?

The best way to prepare inputs is ensuring you have a current findings document and a stored prior state artifact. You must identify the prior baseline using the prior_date metadata field to successfully run the comparison.