scottwater
@scottwater
Public Skills
tracer-autopilot
by scottwater
"Deliver a ticket or spec end to end with per-task implementation and review loops, whole-branch review, and branch finishing."
tracer-code-review
by scottwater
Run Tracer's two-axis Standards-versus-Spec review from a fixed point. Use when /tracer-autopilot calls for its whole-branch review or the user explicitly requests Tracer's Standards and Spec verdicts.
tracer-implement
by scottwater
"Implement a ticket or spec with a visible task plan, sequential task commits, two bounded ticket-wide review passes, and a final test run."
tracer-wat
by scottwater
Ask which skill or flow fits your situation. A router over the tracer skills.
tracer-convince-me
by scottwater
Build the strongest honest, end-to-end case that completed work satisfies its original goal.
code-quorum
by scottwater
Run independent code-review agents and synthesize verified findings.
tracer-interview-me
by scottwater
A relentless interview that asks every frontier question at once, round by round. The entry point of the main flow — interview until shared understanding, then hand off to /tracer-to-spec. Docs-aware in a codebase (reads/writes CONTEXT.md and ADRs), stateless without one.
agent-speak
by scottwater
Writing documents for agents. Use when creating or editing skills, or modifying AGENTS.md or CLAUDE.md.
rigor
by scottwater
Apply a focused, evidence-driven engineering workflow to one concrete task.
blast-radius
by scottwater
Analyze and prove the external breakage risks of a proposed change.
bro
by scottwater
Restate the previous assistant message in plain, concise language.
how
by scottwater
Explain how a subsystem works or where code should live. Use for runtime walkthroughs, architectural ownership, layering questions, and architecture critique; use why for historical motivation.
stop-slop
by scottwater
Rewrite prose to sound direct, readable, and distinctly human. Use when drafting, revising, or reviewing reader-facing writing.
why
by scottwater
Investigate why code or a design decision exists. Use for rationale, rejected alternatives, regressions, postmortems, historical constraints, and evidence-backed thresholds; use how for runtime behavior.
personas
by scottwater
Apply one or more expert lenses to feedback, critique, or guidance.
committing-with-guidelines
by scottwater
Create git commits with atomic scope and product-focused messages. Use whenever preparing, writing, or reorganizing commits.
eli5
by scottwater
Explain a topic like I'm a 5 year old with a simple HTML picture artifact.
tracer-worktrees
by scottwater
Set up an isolated workspace when feature work starts on main/master — reuse any existing feature branch or isolation, prefer the harness's native worktree tool, and fall back to git worktree.
simplify
by scottwater
Refine an implementation by simplifying, deduplicating, reorganizing, and removing low-value tests or unnecessary churn while preserving behavior.
dsa-codebase-audit
by scottwater
Audit a whole codebase for material simplifications in data structures, state, algorithms, control flow, and ownership.
rails-security
by scottwater
"Run a multi-agent security review of Rails application code. Use when the user asks for a Rails security review or audit, or raises a specific concern — authorization/IDOR/tenant isolation, XSS/SQL injection/SSRF, vulnerable gems/Brakeman/bundle audit, or prompt injection in AI features."
tracer-setup
by scottwater
Configure a repo for the tracer skills — set up its issue tracker and domain doc layout. Run once before first use; everything falls back to local files if you skip it.
tracer-prototype
by scottwater
Build a throwaway prototype to answer a design question. Use when the user wants to sanity-check whether a state model or logic feels right, or explore what a UI should look like.
tracer-research
by scottwater
Investigate a focused question against high-trust primary sources and capture cited findings as one Markdown note. Use for documentation, API, specification, or local knowledge-base research, including AFK Wayfinder research tickets.
tracer-to-spec
by scottwater
Turn a settled interview or cleared Wayfinder map into a spec and publish it — no new interview, just synthesis of decisions already made.
tracer-to-tickets
by scottwater
Break a plan, spec, or the current conversation into a set of tracer-bullet tickets, each declaring its blocking edges — one file per ticket locally, or native blocking links on a real tracker.
tracer-wayfinder
by scottwater
Plan an effort too large and foggy for one agent session as a shared map of decision tickets, then resolve them one at a time until the route to a spec is clear.
tracer-codebase-design
by scottwater
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.
tracer-domain-modeling
by scottwater
Build and sharpen a project's domain model. Use when the user wants to pin down domain terminology or a ubiquitous language, record an architectural decision, or when another skill needs to maintain the domain model.
tracer-finish-branch
by scottwater
Finish a development branch — verify tests with fresh evidence, then present merge / PR / keep / discard options and execute the choice, cleaning up worktrees safely.
tracer-handoff
by scottwater
Compact the current conversation into a handoff document for another agent to pick up.
tracer-tdd
by scottwater
Test-driven development. Use when the user wants to build features or fix bugs test-first, mentions "red-green-refactor", or wants integration tests.
github-social
by scottwater
Generate a 1280x640 GitHub repository social preview (Open Graph) image and link it at the top of the README.
gpt-image-prompts
by scottwater
Turn a visual idea into a production-ready prompt and API settings for OpenAI GPT Image models (gpt-image-2).
solo
by scottwater
Use when the user asks to use Solo (MCP) to spawn coding agents, manage scratchpads, todos, or timers, to inspect Solo-managed processes, or when another skill delegates its agent dispatch to Solo.