Claude Skills by simonasrazm

Internal evidence-first troubleshooting engine used by dig-deeper wrappers. Requires a wrapper-supplied authority profile and independent Troubleshooter and Interrogator contexts.

Diagnose nasty, deep bugs when observation is not enough. Use when tools, safe reproduction and controlled experiments are needed to find the cause. Tool installation follows the user's consent or established execution boundary. Diagnosis only, never remediation.

Diagnose nasty, deep bugs and failures using logs, system state, documentation, code and other read-only evidence. Use when the obvious explanation is weak, several causes fit, or earlier fixes treated symptoms. Does not change the target.

Get a moderately better result from a short or underspecified prompt through a lighter, much faster Smart Shot. Use when the task is bounded and reversible, a few direct checks can resolve the important uncertainty, and specialist agents or independent review would cost more than the task deserves.

Find a workable route to a known goal. Use when permissions, policies, unfamiliar systems or technical constraints block the obvious path. Compare real options, test the important uncertainty and hand implementation a supported direction.

Implement or repair one coherent product change with behavior-first evidence, useful seams, and proportional architecture. Use for a scoped change or when another workflow delegates a mutation unit.

Independently verify a frozen candidate through fast, vertical-slice, or final QA and return evidence, defects, boundaries, and useful feedback. Use for QA after product changes, final simulations, or a standalone quality check.

Independently verify visual presentation and interaction in the form people experience them. Use when creating, changing or reviewing human-facing visual output, including configured surfaces and static artifacts.

Reusable control plane for discovering, mapping, coordinating, and resuming a foggy or context-spanning effort. Use when the route itself needs durable decisions, dependencies, frontier state, or Wayfinder-compatible continuation; callers supply domain-specific delivery and verification policy.

Independently assess a change, product, workflow, or deployment for realistic security and privacy risk, reporting outcome separately from coverage. Use for trust-boundary or exposure changes and explicit security review.

Drive a requested outcome through disciplined flow. Can solve complex and compound problems as well as surgical fixes. Use when the user names SFLO Waydriver or wants a problem carried to an evidenced destination, including strategy, research, practical projects, software, and so on.

Make an agent skill cheaper to load without breaking when it activates or how it behaves. Use when a skill has accumulated too much text and the smaller version must earn promotion through routing and runtime tests.

Find unjustified AI-like filler, visual or wording residue, needless abstraction, and process bloat in a candidate; optionally repair bounded presentation-only defects. Use for reviewing interfaces, prose, code, analytics, plans, or explicit slop review.

Discover what an ambiguous or consequential request really requires, including hidden intents, affected domains and unknown unknowns, then assemble ad hoc expert agents, use the right tools, act within authority and independently test the result. Use when the prompt may underspecify the real job or a wrong interpretation could materially change the outcome; deeper and more expensive than Fast Shot.

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

Explain material changes so a person can quickly understand what was solved, where it fits and what needs their attention. Use for a work handover, logical diff, decision update or catch-up across many changes. Supports any domain, with a compact default and deeper reconstruction when needed.