nhouseholder/cursor-laptop-skills

Same behavior; new seams / folders / decoupling. Convert code to clean architecture — separate concerns, increase modularity, reduce coupling.

Review the changes since a fixed point (commit, branch, tag, or merge-base) along two axes: Standards (does the code follow this repo's documented coding standards?) and Spec (does the code match what the originating issue/spec asked for?). Runs both reviews in parallel sub-agents and reports them side by side. Use when the user wants to review a branch, a PR, work-in-progress changes, or asks to "review since X".

Shared vocabulary for designing deep modules. Use when the user wants to design or improve a module's interface, find deepening opportunities, decide where a seam goes, make code more testable or AI-navigable, or when another skill needs the deep-module vocabulary.

Runs a formal 3-round debate between Claude (Proposer) and Gemini 3.1 Pro (Challenger), ending with an AI verdict via backend API. Use when the user says "debate", "debate this topic", "challenge my idea", or wants to adversarially stress-test a position through structured argument.

Diagnosis loop for hard bugs and performance regressions. Use when the user says "diagnose"/"debug this", or reports something broken/throwing/failing/slow.

Build and sharpen a project's domain model. Use when discussing codebase terminology, writing or editing a GLOSSARY.md, or recording or editing an ADR.

ALWAYS ACTIVE on Cloud Agents. Save decisions, conventions, and finished work to the existing Engram MCP (mem_save / mem_session_summary) so other agents can mem_search them. Do not wait for /wrap-up. If the Engram namespace is error, say so.

Effort, time, and complexity estimation before starting a task. Use when a task feels large, when the user asks "how long will this take", or before entering plan mode on a multi-step task. Produces a breakdown with confidence intervals — does NOT start implementation.

Follow your best recommendation — pick one course, execute it, do not interview. Hand off only secrets, GUI login, or a decision you cannot make.

Session start, read-only. Orient from Engram standing decisions plus recent handoffs, git status, and open todos. Never write a handoff.

A relentless interview to sharpen a plan or design.

Grill the user relentlessly about a plan, decision, or idea. Use when the user wants to stress-test their thinking, or uses any 'grill' trigger phrases.

Compact the current conversation into a handoff document for another agent to pick up.

Always-on standing law for every Cloud Agent on any ProjectsHQ repo: honesty about what was run, fail-closed on unknowns, real-money safety for betting picks, and the things a Cloud Agent must never do. Distilled from ~/AGENTS.md, which does not exist inside a container.

Shape output for a reader with ADHD: lead with the next action, number multi-step work, restate state across turns, suppress tangents, give specific time estimates, make wins visible. Invoke with /i-have-adhd; stays on until "stop adhd mode".

Alias of /kg — audit open work, pick the highest-leverage task, implement it, verify, and report the next action.

Keep going — audit open work, pick the highest-leverage task, implement it end-to-end, verify, then immediately take the next highest-leverage step. Do not pause for routine confirmation. Do not stop because one commit, push, or PR landed.

Performance engineer loop — speed, memory, scalability, extra rendering. Find the bottleneck, then ship a measured fix against an explicit target.

Laziest solution that works: YAGNI, stdlib before custom code, native before dependencies, one line before fifty. Levels lite|full|ultra. Use when the user says "ponytail", "be lazy", "yagni", "simplest solution", "do less", or complains about over-engineering or bloat.

Session retrospective — review what was accomplished in this session versus the original goal. Use when closing a session, doing a mid-session check-in on what we did so far, or validating progress. Captures deviations and learning. Not an artifact producer.

ALWAYS ACTIVE on Cloud Agents. Join the tailnet in userspace, reach JobHub on nicholass-imac, fix Cloud git extraheader, use GitHub and Cloudflare APIs already injected. Do not print secrets. Do not wait to be asked.

Test-driven development. Use when the user wants to build features or fix bugs test-first, mentions "red-green-refactor", or wants integration tests.

Turn the current conversation into a spec and publish it to the project tracker: no interview, just synthesis of what you've already discussed.

Break a plan, spec, or the current conversation into a set of tracer-bullet tickets, each declaring its blocking edges, published to the tracker (edges as text in one file per ticket locally, or native blocking links on a real tracker).

Join an unfamiliar product repo: map architecture and data flow, then apply the highest-ROI structural / duplication / perf / maintainability refactors.

Stop. That last message did not land: re-pitch it.

Strategic next-action prioritizer: anchors on the project's north star, scores impact vs effort vs confidence vs decay, filters already-killed work, and audits its own prior recommendations. Triggers on 'whats next', 'what's next', 'what should i do', 'prioritize', 'what now', or /whats-next.

Session end. Write a lean handoff in this project's handoffs/, commit it on the current branch. Do not auto-ship. Save standing memory to Engram MCP.