From blueprint
Break a design into multiple stories with dependency wiring
npx claudepluginhub urso/claudev --plugin blueprintThis skill is limited to using the following tools:
Break a design document into multiple stories, setting up `blocked-by` dependencies between them.
Provides Ktor server patterns for routing DSL, plugins (auth, CORS, serialization), Koin DI, WebSockets, services, and testApplication testing.
Conducts multi-source web research with firecrawl and exa MCPs: searches, scrapes pages, synthesizes cited reports. For deep dives, competitive analysis, tech evaluations, or due diligence.
Provides demand forecasting, safety stock optimization, replenishment planning, and promotional lift estimation for multi-location retailers managing 300-800 SKUs.
Break a design document into multiple stories, setting up blocked-by dependencies between them.
${CLAUDE_PLUGIN_ROOT}/resources/discovery.md${CLAUDE_PLUGIN_ROOT}/resources/design-operations.md${CLAUDE_PLUGIN_ROOT}/resources/story-operations.md${CLAUDE_PLUGIN_ROOT}/scripts/query-stories.sh${CLAUDE_PLUGIN_ROOT}/scripts/set-status.sh${CLAUDE_PLUGIN_ROOT}/scripts/list-workflows.sh${CLAUDE_PLUGIN_ROOT}/scripts/find-doc.sh$ARGUMENTS
Parse for design file path, name, or ID.
!bash ${CLAUDE_PLUGIN_ROOT}/scripts/list-workflows.sh "" story
Read DISCOVERY_GUIDE, DESIGN_OPS, and STORY_OPS for available tools and procedures.
Locate the design based on user input. Read the design document fully.
Load additional context from the design's frontmatter:
depends-on: For each dependency ID, use find-doc.sh to locate and read the dependent design. These provide architectural context and decisions that this design builds on.references: Read each referenced file (paths are relative to git root). These are project files that inform the design.Keep this context available when proposing stories — it helps identify integration points and constraints.
bash QUERY_STORIES --design <design-id>
Show existing stories for this design so we don't create duplicates.
Read the workflow files listed in the pre-computed context above for story guidelines.
Analyze the design's goals, requirements, and technical approach. Propose a set of stories that:
Present the proposed stories as a numbered list:
1. [Story Title] - [one-line description]
blocked-by: (none)
2. [Story Title] - [one-line description]
blocked-by: 1
3. [Story Title] - [one-line description]
blocked-by: 1, 2
Ask user to review, modify, add, or remove stories before creating them.
Follow the procedures in STORY_OPS to create each approved story. Link each to this design via design: frontmatter. Set blocked-by relationships between them.
If the user requests changes to stories after creation (reorder, modify, cancel), follow the relevant procedures in STORY_OPS.
If the design is in draft status, offer to move it to in-progress:
bash SET_STATUS design <design-id> in-progress
Spawn story-review agents in parallel for each created story file. Pass the story file path as the argument to each agent.
Report all created stories with their IDs, titles, dependency graph, and review results. Suggest:
/develop-story to start implementing the first actionable story/design-review if the design hasn't been reviewed yet## Tasks section with checkboxesready for stories with no blockers, draft if they need refinement