From FORSVN Dev
Designs technical blueprints including tech stack, database schema, API design, file structure, and deployment plan for a defined product or feature.
How this skill is triggered — by the user, by Claude, or both
Slash command
/forsvn-dev:architect-system [product or feature to architect][product or feature to architect]This skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Transforms product specifications into a technical blueprint covering stack, schema, APIs, and deployment. Capability metadata (route triggers, prerequisites, load map, artifact contract) lives in [`routing.yaml`](routing.yaml). Methodology and history: [`references/playbook.md`](references/playbook.md).
agents/_template.mdagents/api-agent.mdagents/critic-agent.mdagents/infrastructure-agent.mdagents/integration-agent.mdagents/scaling-agent.mdagents/schema-agent.mdagents/stack-selection-agent.mdreferences/_shared/artifact-contract-template.mdreferences/_shared/before-starting-check.mdreferences/_shared/execution-policy.mdreferences/_shared/manifest-spec.mdreferences/_shared/mode-resolver.mdreferences/_shared/pre-dispatch-protocol.mdreferences/_shared/reviewable-artifact-contract.mdreferences/agent-manifest.mdreferences/anti-patterns.mdreferences/api-patterns.mdreferences/auth-patterns.mdreferences/database-patterns.mdTransforms product specifications into a technical blueprint covering stack, schema, APIs, and deployment. Capability metadata (route triggers, prerequisites, load map, artifact contract) lives in routing.yaml. Methodology and history: references/playbook.md.
Core question: Will this still work at 10x scale with 10x team?
/discover./clean-code./breakdown-tasks./map-user-flow./extract-service.| Artifact | Source | Required? |
|---|---|---|
research/product-context.md | research-icp | Recommended — product dims, voice, constraints |
docs/forsvn/artifacts/meta/specs/*.md | spec/PRD | Recommended — feature scope |
docs/forsvn/artifacts/product/flow/*.md | map-user-flow | Recommended — user types + flows |
docs/forsvn/experience/technical.md | (any skill) | Optional — stack history + constraints |
.forsvn/index/manifest.json | system | Optional — prior runs + downstream task state |
Apply references/_shared/before-starting-check.md. --fast forces single-agent regardless of scope; All 8 Critical Gates fire in every mode.
Run references/_shared/pre-dispatch-protocol.md. Needed dimensions: spec/PRD reference, scale targets (users / RPS / data), constraints (budget / team skills / latency / compliance), deployment context (greenfield / brownfield / migration). Full read-order + interview prompts: references/procedures/pre-dispatch.md.
Before delivering, the critic-agent verifies ALL of these pass:
references/dependency-classification.md.Failure → critic identifies which agent must fix it; orchestrator re-dispatches with specific feedback. Revision-loop handling + 2-round limit: references/anti-patterns.md § "Revision loop".
architecture/system-architecture.md (active); prior runs renamed system-architecture.v[N].md.canonical — top-level folder; edited in place by humans + future runs; team's authoritative architecture record.skill, version, date, status, stack (=product), review_surface (=html — the forsvn-preview review module renders the themed preview while decision_state: pending), decision_state, review_tool, reviewed_at, reviewer, lifecycle, produced_by, provenance. v2 schema in references/_shared/artifact-contract-template.md.breakdown-tasks, review-work, clean-code, forsvn, operator.Full template + section content + version-increment rule: references/report-template.md.
Previous: /discover or /map-user-flow (optional, both sharpen output) | Next: /breakdown-tasks (decomposes into tasks). Cross-stack: reads docs/forsvn/artifacts/meta/sketches/prioritize-*.md.
Re-run triggers: product spec changes significantly, scale requirements change (10x growth), migrating core infrastructure, adding major new integrations.
Multi-agent default (7 agents: stack-selection · infrastructure · schema · api · integration · scaling · critic). Single-agent fallback when mode-resolver downgrades to fast. Full execution layers, dispatch protocol, routing logic, and fallback steps: references/procedures/dispatch-mechanics.md. Per-agent role + file paths: references/agent-manifest.md.
Session execution profile (single-vs-multi): inherit per references/_shared/execution-policy.md.
Read references/anti-patterns.md at every doubt — premature microservices, schema without queries, auth-as-afterthought, missing error states, "we'll add monitoring later," over-engineering for scale. Revision-loop handling + when-to-defer-instead-of-architecting also live there.
/discover, /prioritize, or /map-user-flow first.After delivery: the operator reviews + sets the decision via the forsvn-preview review module (/forsvn:review), decision_state. Approved → dispatch /breakdown-tasks to decompose into tasks. Suggested edits → re-run with feedback. Denied → loop back to /discover or /map-user-flow.
references/procedures/{pre-dispatch, dispatch-mechanics}.mdreferences/playbook.md, references/agent-manifest.md, references/anti-patterns.md, references/report-template.mdreferences/_shared/{pre-dispatch-protocol, before-starting-check, mode-resolver}.mdreferences/pre-dispatch-prompts.md, dependency-classification.mdtech-stack-patterns.md, tech-stack-matrix.md, file-structure-patterns.md, database-patterns.md, api-patterns.md, auth-patterns.md, deployment-patterns.md, failure-modes.md, interaction-edge-cases.md, security-patterns.mdreferences/examples/saas-invoicing-walkthrough.mdnpx claudepluginhub hungv47/meta-skills --plugin forsvn-devTransforms product requirements into executable technical architecture with tech stack selection, system design, deployment strategies, and architecture reviews.
Transforms product requirements into executable technical architecture via phased workflow for system design, tech stack selection, deployment strategy, and review.
Designs system architecture and high-level technical strategy. Use for new systems or subsystems, major refactors, technology selections, system boundaries, and long-term decisions with broad impact.