Guide exploratory planning and codebase investigation for ambiguous requirements.
npx skills add https://github.com/nullmastermind/viber-router --skill spx-plan
Please help me install this Agent Skill. Skill: spx-plan Source: https://github.com/nullmastermind/viber-router/tree/main/.claude/skills/spx-plan Command: npx skills add https://github.com/nullmastermind/viber-router --skill spx-plan
Helps teams and individuals enter a disciplined "explore mode" to think through ambiguous requirements, investigate code, and clarify trade-offs before any implementation begins. It prevents premature coding by enforcing a stance of discovery, surfacing unknowns, and reducing verification failures later in the workflow.
Ask the assistant to "Help me explore and map the requirements for <feature> and identify the files and decision points I should investigate next."
High-intent search queries and answers about installing and using this skill.
Clarifying ambiguous requirements involves entering a discovery stance that restates needs simply, surfaces gaps, and forces concrete choices before coding. This approach uses guided questioning to map unknowns and establish clear trade-offs.
Investigating codebase structure without writing code requires targeted file reads, execution path tracing, and mapping integration points. This method safely explores the repository to identify decision points and patterns without modifying files.
Mapping decision trade-offs uses structured decisioning to present concrete A/B/C options and offer recommendations grounded in existing repository patterns. It drafts verification steps to validate choices against current architecture.
Coordinating research subagents during pre-implementation planning is possible by preparing briefs for specialized agents like research, docs, and UI. These subagents are instructed to report findings only, maintaining a strict no-implementation boundary.
Exploratory planning is not suitable when requirements are fully unambiguous and implementation can proceed immediately. It is designed strictly for discovery sessions and pre-implementation decision-making where unknowns and trade-offs exist.
PostalocityScan repository documentation, tokens, and assets to persist design guidelines to a local AI config file.
xenotaurGenerate structured technical designs for software features using a multi-phase evaluation framework.
dotlab-hqCoordinate web searches and synthesize findings into a cited research report.
royhenengelGenerate spec-driven implementation plans for software development tasks.
