Skip to main content

Issue Seeding

Goal​

Turn the current VRDex planning docs into a GitHub-native build queue.

For each epic:

  • create one parent epic issue
  • create child implementation issues
  • link child issues back to the epic
  • keep acceptance criteria in the epic issue body

Issue sizing rule:

  • prefer larger, independently testable vertical slices over many tiny bookkeeping issues
  • avoid excessive nesting and tracking overhead
  • an issue can be fairly long if it still represents one coherent, shippable chunk of work
  • only split further when a slice becomes hard to reason about, hard to review, or blocked by clear dependencies

Suggested milestone structure​

  • v0.5
  • v1
  • v1.5

Suggested issue template shape​

Title​

  • concise and implementation-facing
  • example: Add claim state and owner authority model

Body sections​

  • problem
  • scope
  • non-goals
  • acceptance criteria
  • dependencies
  • docs to update

Deferred-but-important issue policy​

If something is explicitly out of scope for the current slice but clearly valuable, prefer filing it as a follow-on issue rather than letting it disappear into a non-goals list.

Good examples for VRDex:

  • custom domains
  • slug history redirects
  • SEO optimization for public profiles
  • LLM-assisted moderation and scanning

Suggested first v0.5 issue list​

EPIC-01 Profile foundation​

  • create base profile schema for people and communities
  • add stable slug generation and uniqueness rules
  • add profile type-aware core fields
  • add profile read/write permissions baseline

EPIC-02 Claim and ownership​

  • implement unclaimed vs claimed profile states
  • add Discord claim flow for person profiles
  • add VRChat proof-code verification skeleton
  • add authority handoff from unclaimed to claimed profile

EPIC-03 Public profile experience​

  • build person public profile page
  • build community public profile page
  • add avatar/banner asset support
  • add short bio and about section UI
  • add theme preset and section ordering basics

EPIC-04 Community submissions and trust labeling​

  • build community profile submission flow
  • enforce schema-limited community field set
  • add community-submitted/unverified badges
  • add source attribution display

EPIC-05 Privacy and opt-out controls​

  • add field-level visibility model
  • add profile visibility UI
  • add claimed-owner opt-out flow
  • add moderation path for pre-claim suppression

EPIC-06 Search and discovery basics​

  • implement search across people and communities
  • add trust-aware ranking and cards
  • exclude opt-out entities from public discovery

EPIC-07 Events and profile associations​

  • add event schema and CRUD flows
  • add community event sections and person event views
  • support unclaimed person profiles in event associations
  • add typed media links with generic fallback
  • add optional world linkage field

EPIC-08 Open docs and platform foundation​

  • scaffold Docusaurus docs structure
  • publish product/docs seed pages
  • write API posture doc
  • write self-hosting and infra posture doc

EPIC-18 Software factory and agentic delivery​

  • separate global repo agent policy from local operator preference
  • add repo onboarding skill and docs-backed onboarding flow
  • define review-recycle workflow and trigger model
  • define orchestrator/supervisor loop and resume semantics
  • define verification stack and human validation package expectations
  • define OpenCode task-pool/server direction
  • define repo-level definition of ready
  • choose product analytics and feature-flagging direction
  • define contributor-friendly, agent-compatible contribution workflow
  • choose LLM/agent observability direction

Practical recommendation​

Seed issues in this order:

  1. schema and authority model
  2. public profile pages
  3. claim flow
  4. community submission + trust labels
  5. privacy + opt-out
  6. event core and profile associations
  7. search/discovery
  8. docs foundation
  9. software-factory/meta-loop foundations in parallel where they unblock faster coding

That order keeps the product coherent while still giving you something demoable quickly.

Suggested follow-on issue bucket​

  • add custom domain support for profiles
  • add slug history and redirect behavior
  • improve public profile SEO and metadata strategy
  • add LLM-assisted moderation and abuse scanning layer
  • define agent-consumable VRDex integration kit and MCP roadmap (#73)
  • publish portable VRDex skill for external partner agents
  • prototype standalone VRDex MCP read tools after public API shape stabilizes
  • evaluate optional VRChat MCP bridge tools for cross-context workflows
  • add world discovery and creator attribution lane (#79)
  • define world profile schema and public world page (#84)
  • connect events to worlds and derive active-world views (#81)
  • add Home page Hot Worlds / Active Venues module (#80)
  • add creator commerce links to profiles and world pages (#82)
  • research marketplace API integrations for creator storefronts (#83)