ContractorKeith

Bid-project scope rules for concrete flatwork. Use when starting, triaging, validating, or packaging a commercial concrete flatwork bid.

Whole-division starter rules for CSI Division 03 - Concrete. Use when starting, triaging, validating, or packaging a commercial concrete bid.

Whole-division starter rules for CSI Division 04 - Masonry. Use when starting, triaging, validating, or packaging a commercial masonry bid.

Whole-division starter rules for CSI Division 05 - Metals. Use when starting, triaging, validating, or packaging a commercial metals bid.

Whole-division starter rules for CSI Division 06 - Wood, Plastics, and Composites. Use when starting, triaging, validating, or packaging a commercial carpentry or millwork bid.

Whole-division starter rules for CSI Division 07 - Thermal and Moisture Protection. Use when starting, triaging, validating, or packaging a commercial roofing, waterproofing, insulation, firestopping, or sealants bid.

Whole-division starter rules for CSI Division 08 - Openings. Use when starting, triaging, validating, or packaging a commercial doors, frames, hardware, glazing, or storefront bid.

Whole-division starter rules for CSI Division 09 - Finishes. Use when starting, triaging, validating, or packaging a commercial finishes bid.

Ask which skill or flow fits the current situation — a router over the skills in this collection. Use when the user isn't sure where to start or which skill applies.

Reviews a change since a fixed point on two independent axes: documented repository standards and the originating ticket or spec. Use before committing or merging a ticket, when reviewing a branch or PR, or when asked to review since a base ref.

Provide a shared vocabulary and practical rules for designing deep modules, choosing seams, improving interfaces, and making code easier to test and navigate. Use when designing or restructuring a module, evaluating abstraction depth, or when another skill needs consistent architecture language.

Diagnose hard bugs and performance regressions with a disciplined evidence loop. Use when the user says "diagnose" or "debug this", or reports something broken, throwing, failing, flaky, or slow.

A relentless interview — small batches of related questions — that stress-tests a plan or design before anything gets built, leaving a paper trail of resolved terms (CONTEXT.md) and decisions (ADRs) as it goes. Use when the user wants to pressure-test an idea, sharpen a plan, or says "grill me".

The handoff skill captures the current session in a concise file so a fresh agent can continue without replaying the conversation. Use it when pausing work, switching sessions, or pairing with `/prototype` or `/kodmap`.

Builds one well-defined ticket from its acceptance criteria through tests, review, commit, and ticket closure. Use when the user invokes `/implement` for a GitHub Issue or local markdown ticket that is ready to build.

Plans work too large for one agent session as a shared map of investigation tickets, resolving one decision at a time until the route is clear. Use when an idea has too many unknowns for a normal spec-and-tickets pass, or the user asks to map a large effort.

Guide the learner through a domain's hands-on build exercise. Use this skill when the learner says /build N or is routed here after passing a domain quiz. Reads the domain's build-exercise.md, sets up workspace/domain-N/, walks through phases with progressive hints, and does NOT write code for them.

Teach a specific domain of the Claude Architect course interactively. Use this skill when the learner says /domain N or selects a domain to study. Reads the domain's content.md and teaches one task statement at a time, adapting depth to the learner's experience level.

Display and manage the learner's progress across all domains. Use this skill when the learner says /progress, /domains, or asks about their overall progress. Reads progress/state.json, shows a summary table, calculates readiness, and recommends next actions.

Run a practice quiz for a specific domain of the Claude Architect course. Use this skill when the learner says /quiz N or is routed here after completing domain content. Reads the domain's quiz-questions.md for questions, presents one at a time, then loads quiz-answers.md to reveal answers after each response. Tracks score and identifies weak areas.

Review the learner's build exercise implementation against acceptance criteria. Use this skill when the learner says /review, pastes code, or describes what they built. Evaluates against the build-exercise.md acceptance criteria and gives specific feedback tied to exam concepts.

Create a new skill file and add it to /skills/. Use this whenever a capability gap is identified mid-session — something that would be useful now and in future sessions. Trigger when someone says "we need a skill for X", "can you learn how to do Y", or when a repeated pattern emerges that isn't covered by an existing skill.

Write articles, blog posts, tutorials, and long-form content that reads like a human wrote it. Use this skill whenever the user asks to write an article, blog post, tutorial, guide, essay, or any long-form written content, even if they don't explicitly mention "humanized" or "anti-AI." Also trigger when the user asks to "write something that doesn't sound like AI," improve writing quality, or draft content for publication on a blog, newsletter, or website. This skill applies rigorous anti-AI-detection writing patterns derived from linguistic research to produce prose that is specific, grounded, and stylistically human.