CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

plan-product

Use this skill when the user asks to "plan this product", "define the product", "define product docs", "seed Ubiquitous Language glossary and feature inventory", "define a domain concept", "capture a macro-flow", "plan a feature", "create feature guide", "document a use case", "document the use case", "map actor workflows and alternate flows", "sequence development plan", "dependency-ordered development plan", "order delivery steps", "upgrade planning docs", "migrate model v1 to v2", or run any plan-product workflow. Do NOT use this skill for implementing source code, editing repository files directly, technical slice planning / code review / debugging (route to dev-delivery), repo standards conformance (route to dev-standards), post-implementation learnings (route to dev-learnings), PR/release shipping (route to dev-ship), or plugin/skill authoring (route to plugin-builder).

Quality

79%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/plan-product/SKILL.md
SKILL.md
Quality
Evals
Security

Plan Product

[!IMPORTANT] Scope Directive Unified facade and orchestrator for product planning workflows. Coordinates product definitions, Ubiquitous Language glossaries, domain concepts, feature specifications, actor use cases, phased development plans, and model migrations. Procedural workflows are encapsulated in workflows/; deterministic path resolution, initialization, migration, and validation live under src/ (src/init/init.mjs, src/migration/migrate.mjs, src/validation/validate.mjs).

Out-of-Scope Cross-Plugin Routing Boundaries:

  • Source code implementation, technical phase/step planning (docs/plans/phase-pp-<slug>.md or docs/plans/dp-pp.ss-<slug>.md), causal-chain debugging, and diff/PR code review -> dev-delivery
  • Language/architecture coding standards, check_adoption.py, AST checkers, and docs/code-map/codegraph.db indexing -> dev-standards
  • Post-implementation grounded project learnings (docs/learnings/) -> dev-learnings
  • Verified GitHub PR merges ("push it") and SemVer/tag/marketplace releases ("ship that") -> dev-ship
  • Agent plugin scaffolding, skill authoring, topology reorganization, and multi-host harness-adapter packaging -> plugin-builder

Canonical Numbering & Terminology Contract (v2)

Numbering PatternCanonical NameSegment DefinitionsPath / Location
C-nn
(e.g. C-01)
Concept ID• nn = Concept Number (01–99): Sequential atomic domain model or cross-feature macro-flow (kind: domain-model or kind: flow).docs/concepts/c-nn-short-slug.md
F-gg.nn
(e.g. F-01.02)
Feature ID• gg = Feature Group (00–99): Cohesive functional area / domain clustering related features (00 = Foundation).
• nn = Feature Number (01–99): Sequential feature within that Feature Group.
docs/features/f-gg.nn/feature.md
UC-gg.nn.uu
(e.g. UC-01.02.01)
Use Case ID• gg = Feature Group (00–99): Inherited from parent Feature.
• nn = Feature Number (01–99): Inherited from parent Feature.
• uu = Use Case Number (01–99): Sequential actor workflow within F-gg.nn.
docs/features/f-gg.nn/uc-gg.nn.uu-short-slug.md
DP-pp.ss
(e.g. DP-01.03)
Development Plan Step ID• pp = Phase (00–99 or 01–99): Technical delivery stage in build order.
• ss = Step (01–99): Ordered unit of delivery work within Phase pp that completes/verifies target UC-gg.nn.uus (or F-gg.nn fallback).
docs/plans/development-plan.md

Command Fast-Paths & Triage Matrix

When invoked directly via slash command, argument hint, or explicit user intent, immediately dispatch to the corresponding internal runbook:

User Intent / CommandTarget WorkflowRunbook Path
/plan-product, define a product, target audience, high-level feature inventory, seed/refresh core docs (docs/product.md, docs/GLOSSARY.md, docs/features.md, docs/tech-stack.md)Product Discovery & Inventoryworkflows/product.md
/plan-concept, define a domain concept, entity model, or cross-feature macro-flow (docs/concepts/c-nn-short-slug.md)Concept & Macro-Flow Specificationworkflows/concept.md
/plan-feature, define or split a feature, establish capability boundaries, acceptance checks (docs/features/f-gg.nn/feature.md)Feature Specificationworkflows/feature.md
/plan-use-case, map an actor workflow, entry route, failures, alternate flows, ASCII process tree (docs/features/f-gg.nn/uc-gg.nn.uu-short-slug.md)Use Case Specificationworkflows/use-case.md
/plan-development, sequence delivery steps (DP-pp.ss), define - Verify: checkpoints, order dependencies (docs/plans/development-plan.md)Development Plan Sequencingworkflows/development-plan.md
/plan-maintain, upgrade legacy planning documents (v1 or legacy docs/roadmap.md to v2), reconcile models, or repair referencesMaintenance & Upgradeworkflows/maintain.md

If the user's intent is broad or ambiguous (e.g. "help me plan my app"), start with workflows/product.md.


Core Operating Invariants

  1. Path Authority, 5-Root Taxonomy, & Opt-In Telemetry: Configuration in .nyssa-ai/plan-product/config.json resolves in strict precedence from user_home (~/.nyssa-ai/plan-product/config.json) to repo_root (<repo_root>/.nyssa-ai/plan-product/config.json), falling back to plugin_root (skills/plan-product/assets/config.json), including optional runtimeRoot and telemetry.enabled settings. <repo_root>/.nyssa-ai/plan-product/config.json is the repository path authority; all committed output paths are repo_root-relative. runtime_root resolves via the 4-step precedence contract: (1) --runtime-root -> (2) NYSSA_RUNTIME_ROOT (namespaced to <NYSSA_RUNTIME_ROOT>/.nyssa-ai/plan-product/runtime/ when pointing at a persona workspace) -> (3) active persona cwd containing both SOUL.md and IDENTITY.md (<persona_cwd>/.nyssa-ai/plan-product/runtime/) -> (4) layered config.json runtimeRoot (user_home -> repo_root), <repo_root>/.nyssa-ai/plan-product/runtime/ when <repo_root>/.nyssa-ai/plan-product/ exists, or ~/.nyssa-ai/plan-product/runtime/. Usage telemetry is strictly opt-in ("telemetry": { "enabled": false } by default): zero usage/ directories or files are created unless telemetry.enabled === true, in which case events append to <runtime_root>/usage/events.jsonl (never directly under .nyssa-ai/plan-product/usage/).
  2. Deterministic Bootstrapping: If <repo_root>/.nyssa-ai/plan-product/config.json or required templates are missing, run node "${PLUGIN_ROOT}/src/init/init.mjs" from repo_root before interviewing. Resolve ${PLUGIN_ROOT} (plugin_root) in order of precedence: PLAN_PRODUCT_PLUGIN_ROOT, CLAUDE_PLUGIN_ROOT, CURSOR_PLUGIN_ROOT, PLUGIN_ROOT, or the relative path to this installed plugin directory.
  3. Strict Model v2 Enforcement (Zero Dual v1/v2 Operation): This plugin operates exclusively under Model v2 rules. It CANNOT attempt to operate under the rules of Model v1 and Model v2 simultaneously. If legacy .plan-product/ exists, config.version < 2, legacy docs/roadmap.md or R-pp.ss / R-NNN paths exist, or legacy F-Nxx / UC-nnn paths are detected, stop immediately, alert the user of the version conflict, and dispatch to workflows/maintain.md to run migration (node "${PLUGIN_ROOT}/src/migration/migrate.mjs") before attempting any planning operations.
  4. Preserve User Content: Merge by section heading (Merge-by-Heading Preservation) and preserve unmatched sections, comments, links, and local notes. A file matching its configured template byte-for-byte is template only (Template-Only Detection) and may be initialized without confirmation.
  5. Non-Technical Feature & Use-Case Language: product.md, features.md, feature.md, and uc-gg.nn.uu-short-slug.md MUST be written in plain, non-technical user-outcome and actor-behavior language. Feature titles must follow canonical formulas (Title Case, active verbs, concrete domain artifacts, and user safeguards/protections, e.g. Update Tasks & Projects without Overwriting Personal Notes or Define, Refine & Update Ideal Client Profiles (ICPs)). Plumbing jargon (Surgical, Ingestion, Storage, Syndication), vague 1–2 word stubs (Reminders, Shift Swaps), and user-story phrasing ("Allow users to...") are strictly banned (Canonical is permitted). Never leak overly technical decisions, implementation mechanics, or specific product/technology selections into features or use cases. Technology context belongs in tech-stack.md, and concrete technical decisions or technology selections may appear in docs/plans/development-plan.md entries.
  6. Ubiquitous Language Alignment (docs/GLOSSARY.md): init and /plan-product seed and maintain docs/GLOSSARY.md (<app>/docs/GLOSSARY.md in monorepos). Always read paths.glossary (docs/GLOSSARY.md) before drafting domain concepts, feature titles, actor terms, or development plan steps so product docs enforce canonical domain terms and avoid banned synonyms (ARCH-NAME).
  7. Structured Interview: Ask one blocking question cluster at a time. Present numbered choices in terminal runtimes. Mark inferred items with [inferred] and confirm before writing.

Verification

Before concluding any planning workflow:

  • Confirmed path resolution via .nyssa-ai/plan-product/config.json (user_home -> repo_root).
  • Confirmed template read immediately prior to document write.
  • Confirmed candidate capsule, glossary terms, feature boundary, use case flow, or development plan step sequence with user prior to writing.
  • Verified feature titles adhere to canonical naming formulas without banned plumbing jargon (Surgical, Storage, Ingestion, Syndication), vague stubs, or user-story phrasing.
  • Verified no overly technical decisions or technology selections leaked into feature or use-case documents (kept in tech-stack.md or docs/plans/development-plan.md only).
  • Verified all markdown links are relative to document location and resolve to existing targets.
  • Preserved all unedited user-authored sections and local notes.
  • Executed the target workflow's specific ## Verification Checklist.
  • Reported all created or updated paths to the user.
Repository

NyssaAI/product-planning

Path

skills/plan-product/SKILL.md

Last updated
First committed

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.