knowledge-dungeon

Adds new data fields or entities to Knowledge Dungeon's persistence layer, covering type definitions, store updates, save/load logic, backward compatibility, and test coverage.

Adds a new interactive structure type (building, landmark, or interaction point) to the Knowledge Dungeon village hub. Use this skill when adding any new structure that players can approach and interact with - fishing ponds, new buildings, interactable decorations, or custom landmarks. Covers the full cross-agent pipeline: layout data, Phaser rendering, React info panel, and interaction routing.

Scaffolds a new Phaser game object (NPC, decoration, portal, item, or interactive element) in a Knowledge Dungeon scene, following the project's patterns for texture loading, positioning, animation, and state integration.

Discover which LLM models are available to the user (cloud subscription + local Ollama), classify each generated agent's workload, and recommend (and optionally apply) a per-agent model assignment so that lightweight agents do not default to the most expensive model. Use this skill after `forge-build-agent-team` has produced an agent team, or any time the team changes.

Analyze a Product Requirements Document (PRD), Product Vision with Feature documents, or Feature PRD and generate a complete team of custom agents and reusable skills tailored to the project. Use this skill when asked to create, scaffold, or design a development team from requirements documents.

Build a Feature PRD that captures a feature - either as a new addition to an existing project, or as part of initial project decomposition from a Product Vision. Use this skill when you need a self-contained feature document with user stories, requirements, and implementation tasks.

Build a comprehensive Product Requirements Document (PRD) or Technical Specification from a user's idea, concept, or research document. Use this skill when asked to create, draft, or formalize a PRD, spec, or requirements document.

Decompose an existing monolithic Product Requirements Document (PRD) into a Product Vision document and individual Feature documents. Use this skill when you want to break a large PRD into independent, self-contained features that can be prioritized, built, and delivered separately.

Audit existing skills in .agents/skills/ against agentskills.io best practices. Scores each skill on context economy, gotchas coverage, procedural clarity, progressive disclosure, calibration, and validation. Produces a reviewable audit report and can optionally apply targeted improvements.

Coordinate specialist agents through PRD implementation phases. Provides the full execution process for PRD full builds, Feature PRD executions, and feature-based builds from decomposed documents. Use this skill when the project-orchestrator agent invokes it, or when you need to drive a build step by step.

Applies a unified visual theme (color palette, typography, border styles, and iconography) across both the React UI shell and Phaser game canvas in Knowledge Dungeon, following the research in docs/research/ui-enhancements.txt.

Writes Vitest + React Testing Library tests for a Knowledge Dungeon React UI component, covering render, user interaction, state integration, and edge cases.

Turn a document or plain text into an importable Knowledge Dungeon subject with topic rooms, edges, and notes that include Summary, Key Points, and Recall Question.

Analyze a repository and generate a Knowledge Dungeon subject folder (dungeon.json + room notes/artifacts) for architecture, systems, features, and workflow learning. Use when user asks to build a playable repo mindmap, onboarding dungeon, architecture study map, or phase-ready creator/scribe/archaeologist subject from code.