38 portable Agent Skills for disciplined, reviewable AI-assisted work across Codex, Claude Code, GitHub Copilot, and compatible agents.
nevitonsantana/adaptive-skills
Skill Claude CodeCodex
Evaluate whether an agent can discover, understand, request, execute, confirm, and recover from a meaningful user-owned action with explicit authority and side-effect boundaries.
nevitonsantana/adaptive-skills
Skill Claude CodeCodex
Route broad discovery, SEO, generative visibility, and agent-readiness goals to the smallest sufficient AI Discovery building blocks.
nevitonsantana/adaptive-skills
Skill Claude CodeCodex
Design repeatable, dated evaluations for search discovery, generative retrieval and citation, entity representation, and agent-task outcomes without hiding variance or uncertainty.
nevitonsantana/adaptive-skills
Skill Claude CodeCodex
Evaluate generative retrieval, source selection, citation, answer coverage, and representation fidelity, then form bounded improvement hypotheses with repeatable verification.
nevitonsantana/adaptive-skills
Skill Claude CodeCodex
Review and stabilize how people, organizations, brands, products, methods, claims, and relationships are represented across human-readable and machine-readable surfaces.
nevitonsantana/adaptive-skills
Skill Claude CodeCodex
Evaluate whether search systems can discover, access, crawl, render, index, canonicalize, interpret, and navigate a digital property using available evidence.
nevitonsantana/adaptive-skills
Skill Claude CodeCodex
Review assistant tone, semantic safety, and fact-versus-inference boundaries for crisis-facing responses.
nevitonsantana/adaptive-skills
Skill Claude CodeCodex
Read a crisis context by separating signal from noise, naming stage, and surfacing the next meaningful watchpoints.
nevitonsantana/adaptive-skills
Skill Claude CodeCodex
Design monitoring coverage as an operational sensing system with useful sources, tiers, keywords, noise filters, and readiness.
nevitonsantana/adaptive-skills
Skill Claude CodeCodex
Shape a contract so inputs, outputs, errors, names, and compatibility are clear and hard to misuse.
nevitonsantana/adaptive-skills
Skill Claude CodeCodex
Review boundaries, coupling, complexity, and maintenance cost before consolidating a structural decision.
nevitonsantana/adaptive-skills
Skill Claude CodeCodex
Turn messy research and signals into structured product or business strategy with explicit facts, interpretations, hypotheses, and implications.
nevitonsantana/adaptive-skills
Skill Claude CodeCodex
Insert a short review pause between rounds so work can continue, stop, or hand off with better discipline.
nevitonsantana/adaptive-skills
Skill Claude CodeCodex
Keep code readable, cohesive, and easy to change without turning style into dogma.
nevitonsantana/adaptive-skills
Skill Claude CodeCodex
Explain changes, failures, options, and handoffs with clarity, evidence, and low noise.
nevitonsantana/adaptive-skills
Skill Claude CodeCodex
Reproduce, isolate, fix, and guard against recurrence instead of patching symptoms by instinct.
nevitonsantana/adaptive-skills
Skill Claude CodeCodex
Review an artifact against a declared design system source bundle, classify conformance and candidate patterns, and return source-backed evidence without promotion authority.
nevitonsantana/adaptive-skills
Skill Codex
Plan, write, restructure, and validate self-service technical documentation for products, systems, APIs, libraries, and internal tools. Use when creating onboarding, manuals, tutorials, how-to guides, reference pages, explanations, troubleshooting, FAQs, changelogs, migration paths, or audience-aware documentation…
nevitonsantana/adaptive-skills
Skill Claude CodeCodex
Align project language across product intent, domain concepts, documentation, ADRs, and code before implementation or architectural change.
nevitonsantana/adaptive-skills
Skill Claude CodeCodex
Guide broad experience-design requests to the smallest sufficient strategy, critique, writing, or design-system building block.
nevitonsantana/adaptive-skills
Skill Claude CodeCodex
Estimate the permanent cost of a feature — cognitive, technical, operational, and governance carry — before the build commitment, producing a complexity scorecard and a reduction recommendation.
nevitonsantana/adaptive-skills
Skill Claude CodeCodex
Turn a feature request into a small, testable delivery plan with scope, risks, slices, and proof.
nevitonsantana/adaptive-skills
Skill Claude CodeCodex
Judge whether a proposed feature is worth doing — business intent, revenue or operational lever, user evidence, opportunity-tree fit, complexity cost, and overreach risk — using governed knowledge packs when available and general criteria when not.
nevitonsantana/adaptive-skills
Skill Claude CodeCodex
Close a round of work with enough verified context for the next round without dragging the full session forward.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: