platty

Use when a Figma URL, Figma page, Figma section, or Figma frame must become reusable, revisioned design evidence before Platty MCP product specification or technical design.

Use when creating locally saved MCP-grounded SDD technical design and executable implementation plans from existing prd.md and user_stories.md.

Use when a user separately requests technical or system design from approved product documents with a Figma URL, current-session Figma evidence, or a validated figma_handoff.json discovered beside the product pair.

Use when a Figma URL accompanies a request to organize, draft, create, or improve a product plan, planning document, feature brief, requirements, PRD, user stories, 기획서, 요구사항, or 기능 기획, with or without existing product documents.

Use when creating locally saved MCP-grounded SDD PRD and user-story drafts from a product idea, feature request, PRD need, policy change, or requirements discussion.

Use when a task should use configured Platty MCP tools for remote project context, tool capability checks, Figma-backed product-plan or 기획서 authoring, Figma evidence routing, client setup routing, Platty MCP retrieval, memory or glossary-alias lifecycle routing, or MCP-grounded SDD file creation.

Use when registering, validating, or troubleshooting a Platty MCP endpoint from an MCP-capable runtime such as Codex or Claude Code.

Use when a Platty MCP question asks what changes, what breaks, blast radius, affected screens/APIs/services, implementation locations, cross-EPIC effects, or technical-design impact.

Use when a user explicitly asks to read, record, correct, update, or delete Platty memories or glossary aliases through configured Platty MCP tools.

Use when answering Platty project questions through configured read-only MCP tools, including domain terms, epics, business documents, specs, exact code locations, or source confirmation, or when another Platty MCP skill needs an Impact Seed Packet.

Use when listing, filtering, including, deprecating, or reviewing Platty technical documentation targets before docs generation.

Use when generating, validating, reviewing, resuming, or repairing Platty generated outputs through the public generate-docs workflow.

Use when recording, updating, or removing human knowledge (why, corrections, constraints) anchored to Platty epics, documents, or document items.

Use when Platty CLI and agent skills are already installed and a user wants the first end-to-end project journey from repository registration through static analysis and generated SOT, or wants to resume that multi-phase onboarding journey.

Use when searching local Platty SOT outputs, generated docs, graph traces, code search, or source-grounded project evidence through the full Platty operator plugin.

Use when creating technical SDD design or implementation-task documents from approved SDD request and user-story files using Platty SOT, graph traces, generated specs, and source code.

Use when turning a rough product idea, feature request, PRD need, or requirements discussion into SDD request and user-story documents grounded in Platty SOT evidence.

Use when initializing Platty, creating/selecting/switching projects, adding/listing/managing registered repositories, inspecting setup state, or using the Platty setup workflow hub.

Use when running, resuming, inspecting, or troubleshooting Platty static analysis, project analysis, repository analysis, repo analysis, pipeline stages, analysis runs, run status, cancellation, or next analysis actions.

Use when refreshing existing Platty generated outputs after source, repository, branch, source-root, static-analysis, or static-map changes.

Use when turning a plain product idea into a source-grounded spec via Platty's ultra-spec pipeline — compile an idea into typed registry facts, diff against the canonical SOT, and produce a one-page summary (and optionally system design/tasks/qa). Distinct from platty-sdd-spec (which writes prd.md + user_stories.md); ultra-spec compiles facts into the registry and diffs them against the code's source-of-truth.

Use when any request is about Platty, the Platty CLI, Platty agent skills, project or repository setup, repo/project analysis, static analysis, generated docs, retrieval/search over Platty outputs, epics, business docs, memory, sync, or choosing the right Platty workflow across Codex and Claude Code.

Use when deciding which Platty CLI root command, project workflow, repository workflow, analysis workflow, document workflow, or Platty skill should handle a request.

Use when configuring, starting, validating, or troubleshooting the Platty context backend that exposes a read-only MCP endpoint.