By IniZio
Structured AI development: orchestrated fan-out, advisor gate, intent routing, and skill-driven workflow for Claude Code
Executor-first workflow with advisor checkpoints at hard decisions AND mandatory gate approval before declaring any task complete. The advisor operates as a strategic technical consultant — providing deep architectural insight, trade-off analysis, and effort-aware recommendations — not just a yes/no gate. ALWAYS required before claiming done.
Implementation orchestration skill. Decompose into vertical slices for maximum coder fan-out, validate behavior. Renamed from bdd-implement — loads implement skill.
Use when creating git commits to ensure consistency with repository patterns and minimal commit messages
Merge time-series PRDs and child PRDs into a single time-neutral document that reads as if the system always worked this way. Use after PRD iterations, before handoff, or when cleaning up docs.
Create the master PRD for a feature with enforced filename conventions, standard content format, and session-level mutation tracking. After creation, MUST ask user for review before proceeding to implementation.
Called by the ORCHESTRATOR only — not by executor agents. Gates plan approval and task completion with APPROVE/REVISE/REJECT verdicts. Use for strategic decisions, architecture trade-offs, and as the mandatory final gate before declaring any task complete.
Primary coding specialist — implements features, fixes bugs, writes and edits code across any number of files. The orchestrator should delegate ALL coding work here.
Final quality gate for plans, code, and architecture decisions. The last line of defense before work is committed. Use for review of significant changes, plan validation, and preventing flawed work from shipping. A false approval costs 10-100x more than a false rejection.
Root-cause analysis, regression isolation, stack trace analysis, build error resolution. Use when something is broken and the cause is unclear. READ-ONLY — recommends fixes, never implements.
UI/UX specialist for intentional, polished experiences. Use for styling, responsive layouts, visual consistency, component architecture, animations, and visual polish. Use when users see it and polish matters. 10x better UI/UX than orchestrator. Best with a model strong at visual taste and high reasoning.
Executor-first workflow with advisor checkpoints at hard decisions AND mandatory gate approval before declaring any task complete. The advisor operates as a strategic technical consultant — providing deep architectural insight, trade-off analysis, and effort-aware recommendations — not just a yes/no gate. ALWAYS required before claiming done.
Mid-session codebase architecture review. All exploration runs in isolated subagents so findings never pollute main context. Surfaces shallow modules, coupling smells, and testability gaps as an HTML report. Use between tasks to retrospect on architecture without bloating the session context window.
Alias for `implement`. Retained for backwards compatibility — load `implement` instead.
Merge time-series PRDs and child PRDs into a single time-neutral document that reads as if the system always worked this way. Use after PRD iterations, before handoff, or when cleaning up docs.
Create the master PRD for a feature with enforced filename conventions, modular content format, and session-level mutation tracking. After creation, MUST ask user for review before proceeding to implementation.
No model invocation
Executes directly as bash, bypassing the AI model
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Workflow plugin for AI coding agents providing structured development practices: PRD-driven development, advisor gates, BDD implementation, parallel background tasks, and context management.
opencode.jsonpi install git:github.com/IniZio/groundworkAdd to opencode.json:
{
"plugin": [
"opencode-pty",
"groundwork@git+https://github.com/IniZio/groundwork.git"
]
}
pi install git:github.com/IniZio/groundwork
Restart your agent. Skills auto-discover.
Note: pi-subagents is bundled as a dependency, enabling agent spawning functionality automatically.
| Tool | Purpose |
|---|---|
handoff_session | Create focused continuation prompt |
set_goal | Manage active session goal |
| Skill | Trigger |
|---|---|
use-groundwork | Every session start — core rules, issue-type routing |
interview | Before PRD creation; standalone for small changes |
diagnose | Bugs and regressions |
create-prd | After interviewing for features (≥1 day) |
advisor-gate | Before declaring done |
bdd-implement | After PRD approval or interviewing |
prototype | Design exploration |
goal | Persistent project goal |
When using Pi with pi-subagents, the following agent types are auto-configured:
| Agent | Purpose |
|---|---|
general-purpose | Orchestrator — main workflow coordinator |
advisor | Strategic decisions, architecture, code review |
coder | Fast implementation, tests, build verification |
designer | UI/UX, styling, responsive design |
explorer | Codebase exploration (read-only) |
pnpm install
pnpm test # run tests
pnpm run check # typecheck
src/ — TypeScript sourcesrc/pi.ts — Pi extension entry point.opencode/plugins/groundwork.js — OpenCode plugin entry point.pi/skills/ — Pi skill definitions.pi/agents/ — Agent definitions (auto-installed on session start)npx claudepluginhub inizio/groundwork --plugin groundworkHarness-native ECC operator layer - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
Feature development with code-architect/explorer/reviewer agents, CLAUDE.md audit and session learnings, and Agent Skills creation with eval benchmarking from Anthropic.
Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review
Intelligent draw.io diagramming plugin with AI-powered diagram generation, multi-platform embedding (GitHub, Confluence, Azure DevOps, Notion, Teams, Harness), conditional formatting, live data binding, and MCP server integration for programmatic diagram creation and management.
Access thousands of AI prompts and skills directly in your AI coding assistant. Search prompts, discover skills, save your own, and improve prompts with AI.
Complete developer toolkit for Claude Code