Vibe-App-Wiki

Make sure to use this skill whenever the user mentions "pass the parcel", "parcel mode", "/parcel", "token saving planning", "multi-agent planning", "stateless execution", "clear context", "independent reviewer", or wants to run a highly token-efficient, robust design-and-execution pipeline where state is passed entirely within a .md plan in docs/plans/.

Activate this persona during Phase 7 and Phase 8 (Execution & QA Verification) of a parcel plan to execute codebase edits with absolute surgical precision, manage build/lint environments, and verify runtime stability.

Activate this persona during Phases 1, 2, and 3 (Scoping, Context Gathering, and User Clarification) of a parcel plan to lock down boundaries and eliminate ambiguity.

Activate this persona during code reviews, architectural planning, or specifically during Phase 6 (Senior Dev Hygiene Review) of a parcel plan to ruthlessly audit code quality, security, minimalism, and structural integrity.

Activate this persona during detailed architectural planning, or specifically during Phase 4 (Detailed Execution Plan) of a parcel plan to ruthlessly enforce the Simplicity Ladder, reject speculative scope, and produce surgically precise file-level blueprints.

Activate this persona during scoping, user flow design, or specifically during Phase 5 (Product Owner Review) of a parcel plan to ruthlessly smooth the product vision, user journey, and user experience by eliminating bloat and complexity.

Make sure to use this skill whenever you need to align codebase reality with stated requirements, validate a user or data journey step-by-step, eliminate batch-generation hallucinations, or surface hidden bugs/dead code via a structured True/False evaluation loop.

Make sure to use this skill whenever the user asks to audit, scan, inventory, or catalogue UI elements in the codebase (e.g., "scan all buttons", "audit all inputs", "list all modals"). It dynamically determines search patterns and extraction criteria based on the element type, produces a structured markdown inventory table, and flags design-system anomalies.

Make sure to use this skill whenever the user asks to add a feature, function, upgrade, idea, or task to the backlog. Use it to analyze the request, gather necessary codebase context, and create a comprehensive entry in the project's backlog-index.md and a separate backlog plan file. This skill creates PARKED items only — it does NOT execute or plan execution.

Orchestrates the final project state synchronization, including changelog updates, feature documentation, and cross-reference validation.

Expert Product Manager + Lead Systems Architect skill for creating, evaluating, and updating the 4 strategic artifacts (App Vision, Core Workflow + Product Shape, User Journey, Core Architecture & Data Architecture). Operates as 3 strictly separate actions (Create, Evaluate, Update) and pressure-tests every line for directness, specificity, and non-contradiction.

Make sure to use this skill whenever the user mentions reviewing UI consistency, auditing designs, checking UI rules, verifying frontend compliance, or when modifying/generating UI and frontend code to ensure strict adherence to the project's established design system (DESIGN.md/design-system.md).

Automates the recording of user decisions, feedback, and tribal knowledge to ensure project consistency and long-term learning across all development tasks.

Periodically reviews, de-duplicates, and restructures the project's Knowledge Capture log to keep it concise, themed, and conflict-free.

Make sure to use this skill whenever the user mentions deploying, pushing code, checking production readiness, finalizing a feature, ending a session, or wrapping up work. Use it to scan for architectural drift, security risks, unoptimized queries, and missing CI/CD infrastructure, shifting vibe-coded code to a hardened production-ready state.

Make sure to use this skill whenever the user mentions running tests, executing npm tests, checking lint rules, linting, code formatting, git pushing, pushing to GitHub, or deploying commits to the remote repository. This skill ensures a secure pre-push pipeline by validating tests and linter output prior to any git push.

Use when checking wiki health, detecting broken links, validating frontmatter, finding orphan pages, or auditing ref/ discoverability. Triggers: 'lint wiki', 'check wiki', 'wiki health', 'broken links', 'index drift', 'frontmatter check'. Soft report only — never blocks deploy.

Use when asking a question about the codebase, looking up a concept, or synthesizing an answer from the wiki + ref/ docs. Triggers: 'what do I know about X', 'look up Y', 'find in wiki', 'search docs', 'wiki says'. Read-only — defers writing to knowledge-capture.

Make sure to use this skill whenever the user mentions "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", "terse", or invokes /caveman. Ultra-compressed communication mode cuts token usage ~75% by dropping filler, articles, and pleasantries while keeping full technical accuracy.

Make sure to use this skill whenever you write, review, or refactor code, or when the user asks to implement a feature or fix a bug. Use it to reduce common coding mistakes, avoid overcomplication, make surgical changes, surface assumptions, and define verifiable success criteria.

Make sure to use this skill whenever the user mentions updating global skills, syncing workspace skills to .gemini, pushing workspace skills, starting a session, or when initializing work in this blueprint repository. This skill copies the skills present in the workspace `Skills` folder to the global `.gemini` config skills folder, ensuring the agent's runtime environment has the workspace's version of the skills.

Make sure to use this skill whenever the user mentions updating workspace skills, syncing skills to workspace, pulling global skills, ending a session to save skills, or when wrapping up work in this blueprint repository. This skill scans the skills present in the workspace `Skills` folder and updates them to match their latest versions in the global `.gemini` config skills folder.

Create and iterate on Agent Skills with proper structure, frontmatter, and progressive disclosure. Make sure to use this skill whenever the user asks to build a new skill, create an agentic instruction, generate a SKILL.md, or automate a workflow into a reusable package, even if they don't explicitly use the word "skill".