workbench

Plugin workbench from marketplace kennykankush-skillpack in kennykankush/skillpack. Commands read from the repo's manifest — 12 days ago.

Skills (16)

• bedrock

  Foundation audit mode. Walk a codebase as an adversarial building inspector - stress-test load-bearing logic to bank-grade, limit-test feature flows by actually running them, and file a fragility report backed by runnable repros. Maintains an AUDIT.md ledger at repo root across runs. Use when the us

• decide

  Turn competing options and available evidence into a reasoned recommendation, with the tradeoffs, unresolved assumptions, and conditions for revisiting it. Use when the user asks which direction to take, what to prioritize, whether something is worth doing, to weigh alternatives, or to decide betwee

• devour

  Enter codebase mastery mode before implementation, or refresh an existing map after changes. Use when the user asks to DEVOUR a repo, deeply onboard, map architecture, trace runtime flows, understand blast radius, refresh MAP.md, or establish what changed since the last study. Persists its atlas to

• frontend

  Agentic frontend design framework, with a code path and a Claude Design path. Use for any UI work where how it looks matters - new pages and apps, revamps, redesigns, polishing, or when the user says it looks generic, amateurish, plain, cheap or "not it", asks for a style ("premium", "modern", "clea

• gauntlet

  Run a realistic user journey end to end and sweep-test the real subsystems behind it. Triage the whole journey, then test the risky stations with varied inputs, volume, and independent verification in an isolated reversible environment. Retain useful findings and reproductions; separate observed beh

• isomorph

  Reason about a whole system by mapping it onto a mature, structurally-similar domain that already paid for its mistakes, then read that domain's laws, invariants, and blindspots back onto the system. Use when the user wants the big-picture / first-principles view of an architecture, codebase, produc

• max-prompt

  Turn a vague idea, intuition, frustration, goal, screenshot, rough request, or incomplete brief into a grounded and execution-ready prompt for the right agent or tool. Use across software, debugging, architecture, infrastructure, operations, research, decisions, writing, product/UI, and creative wor

• memory-scriber

  Capture a conversation's essence into the active host's memory directory, and recall it back like a colleague - not a database. Use at the tail end of a substantive session ("log this", "memory-scribe this"), progressively during one ("scribe as we go", "checkpoint this"), or in reverse when the use

• potential

  See what a codebase wants to become and which opportunities deserve attention. Ground possibilities in existing structure, assess who benefits, ongoing burden, opportunity cost, and smaller alternatives. Two modes - open (what could this become) and wish (whether and how the structure can grant a wi

• probe

  Resolve a consequential uncertainty with the smallest useful experiment before committing to a larger direction. Use when the user asks to test an assumption, check feasibility, run a bounded spike, or find out whether an approach works before building it. Define success, failure, limits, and cleanu

• research-report

  Run a deep research dive on any topic and either produce a polished Quarto HTML report (official mode — files written to research/<umbrella>/<title>/) or reply with structured findings inline (scan mode — no files). Use when the user invokes $workbench:research-report, /workbench:research, /workbenc

• scour

  The reality-check hotline. Mid-conversation, leave the closed room and go to the internet to ground what was just said against how the world actually does it. Two faces - verify (we concluded something confidently; pull the load-bearing claims and check each against real fetched sources - confirmed

• skill-advisor

  Recommend which of the user's ALREADY-INSTALLED skills best fits a task or query. Use when the user asks "what skill should I use for X", "which of my skills fits this", "do I have a skill for Y", "what's at my disposal for Z", "recommend a skill from my toolkit", or when you need to pick the right

• skill-distiller

  Distill a successful chat, project workflow, repeated agent behavior, or a way of thinking into a reusable Codex or Claude skill. Use when the user wants to "skillify" a process, capture a reasoning/voice mode that just worked, create a new skill from what just worked, generalize a workflow without

• totality

  The exhaustive-surface cartographer. Map the COMPLETE surface of any object — a resume, a domain, a dataset, a market, a decision space, a codebase's concerns — with PROVABLE coverage instead of vibes. Anchor against official exhaustive taxonomies, rotate enumeration generators, expand every example

• vision

  Keep a project's vision written down and alive. Creates and maintains VISION.md at the repo root - the statement of what the building is trying to be, which bedrock, potential, devour, and any implementing agent read before judging, dreaming, or building. Three moments - birth (carry a warroom-style

Slash commands (2)

• /workbench:research

  Run a deep research dive — produces a polished Quarto HTML report and consolidated notes.md.

• /workbench:scan

  Quick research scan — replies with structured findings inline. No files written.

Subagents (1)

• skill-manager

  Personal skill-management agent. Handles the full lifecycle of Claude Code skills across three mechanisms (plugins, npx skills, skillfish) — install, uninstall, migrate, audit, and discover new skills from marketplaces or GitHub. Use when the user says "install a skill", "add <owner>/<repo>", "migra

Similar plugins