How do I install the Superpowers Sage:architecture Discovery skill?

Run npx skillmds@latest add hekivo/superpowers-sage-architecture-discovery in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Superpowers Sage:architecture Discovery safe to use?

SkillMD's automated safety review verdict for this skill is PASS. Independent scanners report: SkillSpector: PASS, Skill Scanner: PASS. Capability flags: reads secrets. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Superpowers Sage:architecture Discovery?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Superpowers Sage:architecture Discovery free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Superpowers Sage:architecture Discovery?

hekivo (@hekivo) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Superpowers Sage:architecture Discovery by hekivo

Name: Superpowers Sage:architecture Discovery
Rating: 4.666666666666667 (3 reviews)
Author: hekivo

Superpowers Sage:architecture Discovery

Architecture discovery for Sage/Acorn projects — map existing codebase, discover post types, routes, ACF field groups, Livewire components, Service Providers, identify architectural gaps, architecture decision records, ADR, component boundary map, data flow diagram, risk register, implementation sequencing, discovery session, understanding unknown codebases, use before architecting or building new features, output contract, discover-abilities

hekivo Updated Aug 19, 2026 13 repo stars

File contents

Architecture Discovery

Design-first architecture workflow for WordPress + Sage + Acorn + Lando projects.

Announce at start: "I'm using architecture-discovery to produce an approved architecture spec before any implementation plan."

Input

$ARGUMENTS

HARD GATES

Never write implementation code in this skill.
Never invoke superpowers-sage:building before architecture approval and plan generation.
Never skip section-by-section approval.

Procedure (summary)

Full step-by-step procedures for each phase are in references/discovery-procedures.md.

Branch & scope — ensure feature branch exists, normalize scope in one sentence.
Explore context — read config/poet.php, existing blocks, providers, active plans.
Clarifying questions — gather requirements one at a time (UX, editor, integrations, constraints).
Identify components — list with responsibility, inputs/outputs, dependencies, risks.
Parallel probes — dispatch content-modeler, design-extractor (PANORAMIC), sage-reviewer in parallel; synthesize.
Propose approaches — 2-3 options with trade-offs; ask user to choose. Apply AD-2 preset if porting legacy schemas.
Decision graph — classify tasks as parallel/sequential/gated.
Section: Overview — present + ask approval.
Section: Components & data flow — present + ask approval.
Section: Quality strategy — present + ask approval.
Write spec — docs/superpowers/specs/YYYY-MM-DD-<topic>-architecture.md. Commit.
Spec review loop — dispatch sage-reviewer up to 3 times; escalate on persistent failure.
User approval gate — revise until approved.
Handoff — prepare handoff payload; invoke superpowers-sage:plan-generator.

For detailed procedure steps and the AD-2 preset block, read skills/architecture-discovery/references/discovery-procedures.md.

Output Contract

Full output artifact specifications are in references/output-contract.md.

Required outputs:

Approved architecture spec at docs/superpowers/specs/YYYY-MM-DD-<topic>-architecture.md
Review feedback resolved or explicitly waived
Complete handoff payload for superpowers-sage:plan-generator

For spec template and handoff payload format, read skills/architecture-discovery/references/output-contract.md.

Verification

Before completion, confirm all items:

Spec file exists at docs/superpowers/specs/YYYY-MM-DD-<topic>-architecture.md
Spec has user approval recorded in conversation
At least one reviewer loop executed (or explicit user waiver)
Handoff payload is complete and unambiguous

Failure modes

No design reference available: proceed text-first, mark visual assumptions explicitly.
Scope too broad: decompose into sub-projects and complete first sub-scope only.
Repeated reviewer failures: escalate after 3 attempts.
Missing content model signal: re-run superpowers-sage:content-modeler before finalizing spec.

---name: superpowers-sage-architecture-discoverydescription: Architecture discovery for Sage/Acorn projects — map existing codebase, discover post types, routes, ACF field groups, Livewire components, Service Providers, identify architectural gaps, architecture decision records, ADR, component boundary map, data flow diagram, risk register, implementation sequencing, discovery session, understanding unknown codebases, use before architecting or building new features, output contract, discover-abilities---# Architecture DiscoveryDesign-first architecture workflow for WordPress + Sage + Acorn + Lando projects.**Announce at start:** "I'm using architecture-discovery to produce an approved architecture spec before any implementation plan."## Input$ARGUMENTS## HARD GATES- Never write implementation code in this skill.- Never invoke `superpowers-sage:building` before architecture approval and plan generation.- Never skip section-by-section approval.## Procedure (summary)Full step-by-step procedures for each phase are in `references/discovery-procedures.md`.1. **Branch & scope** — ensure feature branch exists, normalize scope in one sentence.2. **Explore context** — read `config/poet.php`, existing blocks, providers, active plans.3. **Clarifying questions** — gather requirements one at a time (UX, editor, integrations, constraints).4. **Identify components** — list with responsibility, inputs/outputs, dependencies, risks.5. **Parallel probes** — dispatch `content-modeler`, `design-extractor` (PANORAMIC), `sage-reviewer` in parallel; synthesize.6. **Propose approaches** — 2-3 options with trade-offs; ask user to choose. Apply AD-2 preset if porting legacy schemas.7. **Decision graph** — classify tasks as parallel/sequential/gated.8. **Section: Overview** — present + ask approval.9. **Section: Components & data flow** — present + ask approval.10. **Section: Quality strategy** — present + ask approval.11. **Write spec** — `docs/superpowers/specs/YYYY-MM-DD-<topic>-architecture.md`. Commit.12. **Spec review loop** — dispatch `sage-reviewer` up to 3 times; escalate on persistent failure.13. **User approval gate** — revise until approved.14. **Handoff** — prepare handoff payload; invoke `superpowers-sage:plan-generator`.> For detailed procedure steps and the AD-2 preset block,> read `skills/architecture-discovery/references/discovery-procedures.md`.## Output ContractFull output artifact specifications are in `references/output-contract.md`.Required outputs:- Approved architecture spec at `docs/superpowers/specs/YYYY-MM-DD-<topic>-architecture.md`- Review feedback resolved or explicitly waived- Complete handoff payload for `superpowers-sage:plan-generator`> For spec template and handoff payload format,> read `skills/architecture-discovery/references/output-contract.md`.## VerificationBefore completion, confirm all items:- Spec file exists at `docs/superpowers/specs/YYYY-MM-DD-<topic>-architecture.md`- Spec has user approval recorded in conversation- At least one reviewer loop executed (or explicit user waiver)- Handoff payload is complete and unambiguous## Failure modes- No design reference available: proceed text-first, mark visual assumptions explicitly.- Scope too broad: decompose into sub-projects and complete first sub-scope only.- Repeated reviewer failures: escalate after 3 attempts.- Missing content model signal: re-run `superpowers-sage:content-modeler` before finalizing spec.

Frequently asked questions