oracle

Fetch pending FAQ items, classify them, and draft answers from repository context.

2|Updated Jan 8, 2026
One-click install
npx skills add https://github.com/matrixshader/matrix-shader --skill oracle-matrixshader
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: oracle
Source: https://github.com/matrixshader/matrix-shader/tree/main/.claude/skills/oracle
Command: npx skills add https://github.com/matrixshader/matrix-shader --skill oracle-matrixshader

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires curl, python3, and includes scripts (resource) components.

What problem does it solve?

Reduces the manual effort and risk of incorrect public answers by automating the triage, drafting, and publishing workflow for pending FAQ submissions while ensuring every publish is human‑approved.

Core Features & Use Cases

  • Fetches pending questions from the MatrixShader support API and builds context from README and repository files.
  • Classifies submissions into CAN ANSWER, NEEDS HUMAN, or JUNK and drafts concise 1-3 sentence answers for those resolvable from code/docs.
  • Executes approved actions via existing scripts to publish, update, dismiss, or delete entries after explicit user confirmation.
  • Use Case: A support or content operator clears an FAQ backlog by approving drafted answers in bulk and publishing them without manually composing each reply.

Quick Start

Use the oracle skill to fetch pending FAQ items, draft answers from the codebase, and present all drafts for your approval.

Frequently Asked Questions about oracle

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

FAQPage Schema
How do I automate drafting answers for pending FAQ submissions from a support API?

Automate drafting answers for pending FAQ submissions by fetching questions from the support API, building context from README and source files, and drafting concise responses for your approval. This triage workflow classifies items as CAN ANSWER, NEEDS HUMAN, or JUNK.

How does automated FAQ triage classify incoming support questions?

Automated FAQ triage classifies incoming support questions by fetching pending items and evaluating them against repository context. Each submission is sorted into CAN ANSWER, NEEDS HUMAN, or JUNK categories to prioritize drafting responses for resolvable items.

Can I bulk publish drafted FAQ answers without manual review?

You cannot bulk publish drafted FAQ answers without manual review. The workflow requires explicit user approval before publishing, updating, dismissing, or deleting any entries to ensure every public answer is human-approved and accurate.

Do I need curl and python3 to fetch and triage support inbox items?

You need curl and python3 installed to fetch and triage support inbox items. The skill uses these dependencies to execute provided scripts for API actions, fetch pending questions, and process repository files for context.

How do I secure API requests when fetching pending FAQ items?

Secure API requests when fetching pending FAQ items by using Bearer authentication via the DASHBOARD_PASSWORD environment variable. This authentication is required for interacting with the support API to retrieve submissions.

What is the best way to clear an FAQ backlog using codebase context?

The best way to clear an FAQ backlog using codebase context is fetching pending items, building context from README and source files, drafting concise answers, and approving them in bulk. This automates triage while ensuring human approval before publishing.