Product & Planning

marco-souza Skill

Facilitate structured brainstorming sessions to generate, explore, and refine ideas before committing to requirements. Use when the user has a vague concept, wants to explore possibilities, needs creative solutions, or is in the early ideation phase. Do NOT use when the requirements are already clear and the user is ready to plan implementation (use create-prd or prd-to-tasks instead).

marco-souza Skill

Create a Product Requirements Document (PRD) from recent discussions, conversations, or accumulated context. Use when the user wants to formalize requirements, document what was discussed, or create a spec to guide implementation. Do NOT use for technical implementation specs (use SPEC.md from project-files) or for already-implemented features (use CHANGELOG.md).

marco-souza Skill

Convert a Product Requirements Document (PRD) into a structured tasks.json file with tasks, dependencies, priorities, and estimated effort. Use when a PRD exists and the user wants to break it down into implementable tasks, or says "turn this into tasks," "create issues from the PRD," or "what are the next steps?" Do NOT use when there is no PRD or when the user wants to skip planning and start coding immediately.

microsoft Skill

Domain patterns for peer-to-peer system architecture — network topologies, NAT traversal, peer discovery, data distribution, consistency without central authority, identity and trust, and incentive mechanisms. Use when designing or evaluating a decentralized, peer-to-peer, or mesh-networked system.

okwinds Bundle

Write complete, unambiguous PRDs that development teams can implement without guesswork. Includes requirement discovery framework, structured documentation methodology, completeness checklists, and common pitfall avoidance. Use when: writing new PRDs, reviewing PRD drafts, validating requirement completeness, preparing for engineering handoff. Triggers: 'write PRD', '写PRD', '产品需求文档', '需求文档', '需求规格', '需求评审', '完善需求', 'create requirements doc', 'product requirements', 'feature spec', 'requirements document'. Anti-triggers: 'technical design doc', 'architecture design', 'implementation plan', 'API design', '架构设计', '技术方案', '实现方案', '接口设计'.

okwinds Bundle

从产品 PRD 产出“复刻级可落地”的 UI/UX 研发规格文档包（目录同构骨架、公共基座、组件/页面契约、覆盖映射、索引与 worklog）。适用于需要把 PRD 转成前端可复刻实现的规格文档、UI/UX 研发规格、界面契约与验收标准的场景；避免用于只要视觉灵感/纯 UI 赏析或直接写代码实现的请求。

okwinds Bundle

Transform PRD (Product Requirements Document) into actionable engineering specifications. Creates detailed technical specs that developers can implement step-by-step without ambiguity. Covers data modeling, API design, business logic, security architecture, deployment, and agent system design. Use when: converting product requirements to technical specs, validating PRD completeness, planning technical implementation, creating task breakdowns, or defining test specifications. Triggers: 'PRD to spec', 'convert requirements', 'technical spec from PRD', 'engineering doc from requirements', 'validate PRD'.

stanislavnianko Bundle

[discovery-phase pack · synthesis] Builds 2-4 evidence-backed user personas from interview notes, SME workshops, support data, and secondary research. Forces every trait to cite a source — no invented demographics. Produces personas.md. Reads discovery-context.md, themes.md, interview-notes/, sme-notes/, support-data-analysis.md.

stanislavnianko Bundle

[discovery-phase pack · deliverables] Generates a client-facing proposal document consolidating problem understanding, recommended approach, scope, assumptions, dependencies, team, timeline, and headline pricing guidance. Designed for pre-sale or paid-discovery conclusion. Produces proposal.md. Reads discovery-context.md and most upstream artifacts.

stanislavnianko Bundle

[discovery-phase pack · deliverables] Drafts a Statement of Work for legal/procurement review post-proposal acceptance. Captures deliverables, acceptance criteria, payment milestones, change-request process, warranties, and termination clauses. NOT a substitute for legal review. Produces sow-draft.md.

stanislavnianko Bundle

[discovery-phase pack · scoping] Translates a defined scope into an effort estimate broken down by discipline (Dev / QA / BA / PM / Design) × phase, with confidence intervals and explicit assumptions. Designed to feed proposal and sow-draft. Produces estimation.md. Reads discovery-context.md and scope-doc.md.

stanislavnianko Bundle

[discovery-phase pack · deliverables] Produces a single GO / PIVOT / KILL memo consolidating all evidence into a one-page decision document. The only artifact most senior stakeholders read. Adapted for outsourcing — decision-owner is typically the client, agency provides recommendation with evidence chain.

stanislavnianko Bundle

[discovery-phase pack · evidence] Structured workshop sessions with the client's subject-matter experts when end users are unreachable. Captures their view of user pain, current process, edge cases, and existing workarounds — flagged as proxy evidence (lower weight than direct user data). Produces sme-notes/*.md.

stanislavnianko Bundle

[discovery-phase pack · scoping] Defines a Minimum Viable Product as a learning vehicle, not a feature-stripped product. Distinguishes MVP scope from PoC scope and from full-product scope. Often run after feature-scoping when the client conflates "MVP" with "first release". Produces mvp-definition.md.

stanislavnianko Bundle

[discovery-phase pack · validation] Plans a runnable prototype that retires the top assumptions in a realistic environment. Differs from MVP by being internally-scoped (limited audience, throwaway by default). Includes demo script. Produces poc-plan.md and a runnable skeleton scaffold. Reads discovery-context.md.

stanislavnianko Bundle

[discovery-phase pack · scoping] Translates the chosen solution direction and top risks into a tight, build-able scope with explicit in/out boundaries per area. Designed to be reusable as a chunk of a proposal or SoW. Produces scope-doc.md. Reads discovery-context.md, opportunity-tree.md, and risk-assumption-map.md.

stanislavnianko Bundle

[discovery-phase pack · synthesis] Maps the end-to-end journey for a named persona — stages, actions, touchpoints, pain points, emotion line, and opportunity moments. Distinguishes current-state from future-state. Every pain point cites evidence; opportunities link back to opportunity-tree. Produces journey-map.md. Reads personas.md, themes.md, interview-notes/.

stanislavnianko Bundle

[discovery-phase pack · discovery] Restates the client's problem, target user, current workaround, success signal, and renders a falsifiable hypothesis. Pressure-tests against falsifiability, specificity, novelty, and strategic alignment. Produces problem-canvas.md. Reads discovery-context.md.

stanislavnianko Bundle

[discovery-phase pack · foundation] Interviews the BA about the client, the product/initiative, the engagement model, access constraints, and expected deliverable. Writes a single discovery-context.md to the project root. Every other skill in this pack recommends reading it; running it first gives the best results, but other skills will offer inline bootstrap if you skip ahead. TRIGGER when the user says "build my discovery profile", "set up discovery context", "start a new client engagement", or activates any other discovery-phase skill before this one has run.

stanislavnianko Bundle

[discovery-phase pack · evidence] Runs direct end-user interviews to saturation. Warns (does not halt) if discovery-context says user_access is none or proxy-only — recommends sme-workshops / support-data-analysis / secondary-research instead, but allows override. Captures past behavior, not hypotheticals. Produces one note file per participant + saturation log.

stanislavnianko Bundle

[discovery-phase pack · evidence] Maps direct competitors, indirect substitutes, and adjacent (non-software) workarounds for the problem in problem-canvas.md. Identifies the actual gap and any "free incumbent" the client can't outcompete. Produces competitive-scan.md. Reads discovery-context.md.

stanislavnianko Bundle

[discovery-phase pack · deliverables] Hands discovery findings to a delivery team (could be same agency squad in delivery phase, different agency, or client internal team). Captures decisions made, open questions, risks transferred, evidence chain. The "what we know vs what we don't" document. Produces discovery-handoff.md.

stanislavnianko Bundle

[discovery-phase pack · validation] Time-boxed engineering investigation targeting the top feasibility assumptions from risk-assumption-map. Produces tech-spike-report.md and disposable code. Warns (does not halt) if no engineering capacity is reflected in the engagement context — output is tagged speculative if the BA chooses to proceed solo.

stanislavnianko Bundle

[discovery-phase pack · synthesis] Consolidates whatever evidence was actually gathered (interview notes, SME workshops, support data, secondary research, competitive scan) into ranked themes and top-3 pain points with confidence-weighted scoring. Produces insight-matrix.md and themes.md. Reads discovery-context.md.

stanislavnianko Bundle

[discovery-phase pack · discovery] Designs the evidence-gathering plan based on what access is actually available. Branches into direct interviews, SME workshops, secondary research, support-data analysis, or a mix. Produces research-plan.md and (if applicable) interview-guide.md. Reads discovery-context.md.

stanislavnianko Bundle

[discovery-phase pack · evidence] Synthesizes published industry reports, benchmarks, regulatory filings, academic research, and analogous-domain case studies to substitute or augment direct evidence. Especially for engagements where user_access is none. Produces secondary-research.md. Reads discovery-context.md.

stanislavnianko Bundle

[discovery-phase pack · foundation] Optional orchestrator for the discovery-phase pack. Walks the BA through a recommended skill sequence based on engagement type and access level — but does not enforce. Acts as a guide for users new to the pack. Skills can be (and often are) run independently of this conductor. Activate when the user wants a guided sequence rather than picking skills à la carte.

stanislavnianko Bundle

[discovery-phase pack · synthesis] Builds a Teresa Torres-style opportunity-solution tree from the top-3 pain points. If the client pre-specified a solution, this skill validates that solution against alternatives rather than generating cold. Produces opportunity-tree.md. Reads discovery-context.md and insight-matrix.md.

stanislavnianko Bundle

[discovery-phase pack · discovery] Maps client-side AND agency-side stakeholders with their stakes, success metrics, concerns, decision rights, and conflicts. Two-tier map. Surfaces deal-breakers in week 1, not week 4. Produces stakeholder-map.md. Reads discovery-context.md.

stanislavnianko Bundle

[discovery-phase pack · evidence] Mines client's existing data — support tickets, call recordings, chat logs, NPS verbatims, sales call transcripts, product analytics — for unsolicited user signal. Faster and more honest than interviews. Produces support-data-analysis.md. Reads discovery-context.md.

stanislavnianko Bundle

[discovery-phase pack · synthesis] Surfaces implicit assumptions across desirability, viability, feasibility, usability, and ethical/legal axes. Ranks by impact-if-wrong × evidence-today. Names top 3 to retire before scoping. In outsourcing, also distinguishes client-owned vs agency-owned vs shared risks. Produces risk-assumption-map.md.

vfarcic Skill

Run a PRD end-to-end autonomously — start, iterate until done, create a PR, and wait for its CI + bot reviews to settle before reporting. Stops before merge for manual validation.

vfarcic Skill

Take committed work from a branch to a verified pull request — push, open the PR, settle CI and the automated review, answer and resolve every finding, and hand off. Never merges; arms auto-merge only when nothing downstream gates the merge. Use when work on a branch is finished and needs to become a reviewed PR, whether or not a PRD started it.

vfarcic Skill

Build the queue of open PRDs that are actually available to work — those carrying a PRD document and not already in flight — then claim them and dispatch one isolated unit per PRD. Asks the shape per PRD and composes the task DIFFERENTLY for each: a single agent is pointed at /prd-full, while a team is governed by the orchestrator role template and told explicitly not to run it. Use when asked to find a PRD to work on, pick a PRD off the backlog, or run several PRDs in parallel. It does no implementing itself — for one PRD you intend to run yourself, use /prd-full directly.

vfarcic Skill

Build the queue of open issues that are actually available to work — excluding PRDs, anything already in flight, and duplicates — then assign them and dispatch one isolated agent per issue. Asks how many to take, verifies each candidate against origin/main rather than the local checkout, and composes a self-contained task carrying this repo's gates. Use when asked to find issues to work on, pick something off the backlog, or work through several issues in parallel. It does no implementing itself — for one named issue, just work it directly.

vfarcic Skill

Build the queue of open PRs where the ball is in your court — yours needing feedback addressed, others' needing verification — and dispatch one isolated agent per PR to move it toward resolution. Asks how many to take, composes a self-contained task with a per-PR risk note, and keys push permission to authorship. Use when asked to work through the PR backlog, review the open PRs, find what is waiting on you, or clear review feedback across several PRs. It does no verifying itself — for one named PR, use /verify-pr directly.