rye-knowledge-reader

Inspect Rye knowledge, tasks, candidates, and provenance with read-only database access.

2|Updated Feb 20, 2026
One-click install
npx skills add https://github.com/BLTGV/project-rye --skill rye-knowledge-reader
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: rye-knowledge-reader
Source: https://github.com/BLTGV/project-rye/tree/main/skills/rye-knowledge-reader
Command: npx skills add https://github.com/BLTGV/project-rye --skill rye-knowledge-reader

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill eliminates the risk of accidentally modifying Rye's database when you need to inspect its current knowledge, historical state, open tasks, pending review candidates, or source provenance, enabling safe analysis and context building for Rye-assisted workflows.

Core Features & Use Cases

  • Read-Only Knowledge Access: Query current valid assertions, active graph edges, and task statuses without any risk of durable writes to the database.
  • Temporal and Provenance Analysis: Retrieve point-in-time knowledge states, trace how accepted facts link back to source evidence, and clearly distinguish accepted knowledge from pending candidates and raw evidence.
  • Safe Context Building: Use for status checks, knowledge summarization, and audit trail reviews without altering Rye's data. For example, if you need to verify which source documents support the accepted facts about a client and check their open onboarding tasks, this skill lets you pull that information safely.

Quick Start

Use the rye-knowledge-reader skill to pull all current accepted knowledge and open task status for the "First Scope" onboarding scope without modifying the database.

Frequently Asked Questions about rye-knowledge-reader

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

FAQPage Schema
How do I query Rye knowledge graphs without accidentally modifying the database?

You can safely query Rye knowledge graphs without database writes by using a read-only access tool. This skill restricts all operations to inspection only, preventing any accidental data modifications while pulling current assertions and active graph edges.

What is the best way to trace provenance links for accepted facts in a knowledge base?

Tracing provenance links in a knowledge base requires retrieving point-in-time states and following source evidence connections. This skill safely queries accepted facts and maps them back to their original source documents for audit or context building.

Can I check candidate review queues and open task statuses without altering Rye data?

Yes, you can inspect candidate review queues and open task statuses without altering Rye data. The skill enforces a strict read-only policy with RLS session context configuration to retrieve pending candidates and active task statuses safely.

How do I retrieve point-in-time knowledge states for temporal analysis in Rye?

Retrieving point-in-time knowledge states in Rye involves querying historical assertions and provenance links without executing database writes. This skill enables temporal analysis by fetching the exact knowledge state at specific moments safely.

Does read-only knowledge retrieval work for audit trail reviews in Rye-assisted workflows?

Read-only knowledge retrieval works effectively for audit trail reviews in Rye-assisted workflows. It allows you to pull accepted knowledge, source provenance, and task statuses to verify evidence without risking durable writes to the database.

What are the limitations of using a read-only approach for Rye knowledge base inspection?

The limitation of a read-only approach for Rye knowledge base inspection is that you cannot create, update, or delete assertions, tasks, or candidates. It is strictly designed for safe querying, status checks, and evidence tracing without modifying the underlying data.