discover-artifacts

Discover agent-published artifacts by scanning .os/memory directories and parsing YAML frontmatter.

4|Updated Feb 1, 2026
One-click install
npx skills add https://github.com/mkalkere/agent-coordinator --skill discover-artifacts
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: discover-artifacts
Source: https://github.com/mkalkere/agent-coordinator/tree/main/.os/skills/core/discover-artifacts
Command: npx skills add https://github.com/mkalkere/agent-coordinator --skill discover-artifacts

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Find and read work products published by other agents. Artifacts are the primary way agents share results — reviews, analyses, reports, and decisions.

Core Features & Use Cases

  • Discover artifacts stored in memory directories by type (review, analysis, report, summary, decision).
  • Read artifact metadata from YAML frontmatter (type, subject, created_by, created_at, related_to).
  • Use results to avoid duplicating work and build on prior analyses.

Quick Start

List the most recent artifact of a given type from the memory store.

Frequently Asked Questions about discover-artifacts

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

FAQPage Schema
How do I find artifacts published by agents in memory stores?

To find agent artifacts in memory stores, parse YAML frontmatter and scan the .os/memory/{type}/ directories to return structured results like reviews, analyses, reports, and decisions.

What metadata is available when discovering agent reviews and analyses?

When discovering agent reviews and analyses, the available metadata parsed from YAML frontmatter includes type, subject, created_by, created_at, and related_to fields for structured filtering.

Can I search historical project contexts for agent reports and summaries?

Yes, you can search historical project contexts for agent reports and summaries by scanning memory directories to locate prior decisions and avoid duplicating work across active projects.

What's the best way to read prior agent decisions to avoid duplicating work?

The best way to read prior agent decisions and avoid duplicating work is to scan the .os/memory directories by type and parse the YAML frontmatter to build on existing analyses.

Do I need YAML frontmatter to locate agent work products in memory directories?

Yes, YAML frontmatter is required to locate agent work products in memory directories, as it contains the essential metadata fields needed to filter and return structured search results.