Jonathan Gabor

INSPEKTERA (Integrity Navigation: Systematic Pattern Evaluation, Knowledge Tracing; Examine, Report, Advise). ALWAYS use this skill for codebase health audits, architecture reviews, and structural quality assessments. This skill is REQUIRED whenever the user wants to assess codebase health, detect architecture drift, find pattern inconsistencies, identify complexity hotspots, evaluate test coverage quality, or check dependency health. Do NOT attempt codebase-wide quality assessments without this skill because it contains the critical workflow for multi-dimensional evaluation, evidence-based findings, confidence scoring, and trend tracking that prevents noisy or superficial audits. Trigger on: "inspektera", "audit the codebase", "check code health", "architecture review", "find technical debt", "assess code quality", "how healthy is this codebase", "what needs fixing", "structural review", "pattern audit", "dependency check", "test coverage audit", or when realisera has run 5+ cycles without a health check.

DOKUMENTERA: Documentation Origin, Knowledge Unified, Methodology Enforced, Notation Traced. Examine, Record, Articulate. ALWAYS use this skill for creating, updating, or auditing project documentation. This skill is REQUIRED whenever the user wants to write documentation before building a feature (DTC-first), generate docs for existing code, update stale documentation, audit docs against implementation, create README/CLAUDE.md/ AGENTS.md, write API documentation, or maintain project-level documentation. Do NOT create or modify project documentation without this skill. It contains the critical workflow for intent-first documentation, context-detected approach, doc-vs-code verification, and DOCS.md coverage tracking. Trigger on: "dokumentera", "write docs", "document this", "update the docs", "create README", "write CLAUDE.md", "write AGENTS.md", "docs first", "document before building", "audit docs", "check documentation", "docs out of sync", or when planera's plan includes documentation tasks.

INSPIRERA (Insight Navigation: Source Pattern Identification and Resonance, Evaluate, Reframe, Assimilate). Analyzes an external link (GitHub repo, article, blog post, docs page, HN thread) and maps its concepts, patterns, and primitives to one of the user's own projects. Use this skill whenever the user drops a URL alongside a project name or repo link, or asks something like "could I use this in X?", "how does this apply to my project?", "is this relevant to Y?", "what can I take from this?", "what do you think of this library?", "should I adopt X?", "review this repo for ideas", "anything worth stealing from X?", or "how does X compare to what I'm doing?". Also trigger when the user pastes a link and mentions one of their own repos or codebases in the same message, or when they paste a link while working in a local project directory and the context implies they want to apply the ideas. Produces a structured markdown analysis with applicability matrix and recommended next steps.

OPTIMERA (Objective Pursuit: Targeted Iterative Measurement; Experiment, Record, Advance). ALWAYS use this skill for metric-driven optimization of a measurable objective. This skill is REQUIRED whenever the user wants to improve a concrete, quantifiable property of their codebase: test pass rate, benchmark performance, bundle size, latency, lint score, type coverage, or any other metric that can be measured by running a command. Do NOT attempt iterative optimization without this skill because it contains the critical workflow for objective-driven experimentation, eval harness design, structured keep/discard decisions, and safety rails that prevent regressions. Trigger on: "optimera", "optimize", "improve performance", "reduce latency", "increase test coverage", "lower bundle size", "speed up", "make faster", "make smaller", "get the score up", "hit the target", "improve the metric", "benchmark and iterate", "run experiments", "tune", "experiment until", or setting up /loop for recurring optimization.

ORKESTRERA (Orchestration Runtime: Knowledge-coordinated Execution Strategy, Targeted Routing; Evaluate, Resolve, Adapt). ALWAYS use this skill for multi-cycle plan execution, orchestrated autonomous development, and skill-agnostic task dispatch. This skill is REQUIRED whenever the user wants to execute an entire plan autonomously, chain multiple skills together, run multi-cycle development without /loop, or have tasks evaluated by a separate agent. Do NOT attempt multi-skill orchestration without this skill because it contains the critical conductor protocol for plan-driven dispatch, inspektera-gated evaluation, retry handling, and context-lean multi-cycle execution that prevents wasted compute and ensures quality. Trigger on: "orkestrera", "orchestrate", "run the plan", "execute the plan", "run all tasks", "dispatch skills", "multi-cycle", "autonomous plan execution", "keep going until done", "run through the plan", "build everything", or any request for evaluated multi-cycle development.

PLANERA (Planning Logic: Adaptive Notation, Executable Requirements Architecture, Explore, Refine, Articulate). ALWAYS use this skill for structured planning of complex work before execution. This skill is REQUIRED whenever the work is too large for a single realisera cycle, involves multi-file changes across modules, requires task decomposition and ordering, or would benefit from explicit acceptance criteria before implementation begins. Do NOT attempt multi-cycle development without this skill. It contains the critical workflow for scale-adaptive planning, behavioral acceptance criteria, adversarial review, and plan-to- cycle integration that prevents wasted work on complex tasks. Trigger on: "planera", "plan this", "write a plan", "break this down", "decompose this", "how should we build this", "spec this out", "plan before building", "multi-step feature", "this is too big for one cycle", or when realisera encounters work too large for a single cycle.

PROFILERA: Persona Reconstruction, Observable Footprint Indexing Logic. Examine, Reconcile, Articulate. Mines Claude Code session history, memory files, project configs, and conversation data to generate an agent-consumable decision profile with per-entry confidence scoring and dormancy decay. Two modes: Full (regenerate from scratch) and Validate (incremental check of existing profile). This skill should be used when the user says "build decision profile", "generate decision profile", "update my profile", "refresh my decision profile", "rebuild my profile", "regenerate profile", "what would I decide", "analyze my decisions", "mine my sessions", "decision patterns", "review my decision history", "validate my profile", "check my profile", "quick profile check", or "/profilera". Also applies when the user wants to understand their own decision-making patterns across sessions, asks someone else to predict their decisions, or wants a document that captures how they think.

REALISERA (Relentless Execution: Autonomous Loops Iterating Software; Evolve, Refine, Adapt). ALWAYS use this skill for autonomous or continuous development of a project. This skill is REQUIRED whenever the user wants you to independently decide what to build and build it, evolve a project over time, run development cycles, or make autonomous progress on a codebase. Do NOT attempt autonomous development without this skill because it contains the critical workflow for vision-driven development, persona-grounded decisions, structured cycles, and safety rails that prevent wasted work. Trigger on: "realisera", "run a dev cycle", "evolve the project", "develop autonomously", "build the next feature", "keep building", "start building", "work on the project", "refine the vision", or setting up /loop for recurring development. Also trigger when the user has a codebase and wants you to independently decide what to work on.

RESONERA: Reflective Engagement, Socratic Observation Nexus. Explore, Reframe, Articulate. ALWAYS use this skill for structured deliberation before consequential decisions. This skill is REQUIRED whenever the user wants to think through a complex decision, evaluate tradeoffs, scope work before building, choose between competing approaches, or reason through an architectural or strategic choice before committing to it. Do NOT attempt to resolve ambiguous, high-stakes, or multi-faceted decisions by jumping straight to implementation. Use this skill to deliberate first. Trigger on: "resonera", "help me think through", "should I", "I'm torn between", "what's the right approach", "let's reason about", "help me decide", "think this through with me", "let's deliberate", "what are the tradeoffs", "scope this out", "before I build this", "talk me through", or when the user is stuck between options. Also trigger when realisera or optimera brainstorm sessions surface decisions too complex for inline resolution.

VISIONERA: Visionary Inception, Strategic Imagination, Observation Nexus. Explore, Refine, Articulate. ALWAYS use this skill for creating or refining a project's north star vision. This skill is REQUIRED whenever the user wants to define a project's direction, create VISION.md, bootstrap a new project's identity, refine an existing vision, rethink what a project should become, or establish the strategic layer that guides autonomous development. Do NOT create VISION.md without this skill when it is installed. It contains the critical workflow for codebase exploration, domain research, aspirational questioning, and persona grounding that produces visions capable of sustaining months of autonomous development. Trigger on: "visionera", "create a vision", "write VISION.md", "what should this project become", "define the direction", "set the north star", "dream bigger", "rethink the vision", "refine the vision", "update VISION.md", "bootstrap the project", or when realisera detects no VISION.md.

VISUALISERA (Visual Identity: Systematic Unified Aesthetic Language, Intent-driven Style Engineering; Record, Articulate). ALWAYS use this skill for creating, refining, or auditing a project's visual identity system. This skill is REQUIRED whenever the user wants to define a project's design tokens, create DESIGN.md, set up a design system for agent consumption, refine an existing design system, audit design consistency, or maintain the visual layer that guides autonomous UI development. Do NOT create DESIGN.md without this skill when it is installed. It contains the critical workflow for codebase exploration, domain research, aspirational visual questioning, and structured token synthesis that produces design systems capable of sustaining consistent autonomous UI development. Trigger on: "visualisera", "create design system", "write DESIGN.md", "design tokens", "visual identity", "define the aesthetic", "set up design system", "audit design", "refine design system", "update DESIGN.md".

HEJ (Holistic Entry Junction, Orient, Route, Activate). ALWAYS use this skill as the single point of entry to the agentera suite. This skill is REQUIRED whenever the user starts a session, returns to a project, or needs orientation on what to do next. It detects whether the project is fresh (no state artifacts) or returning (existing artifacts), delivers a situational briefing, and routes to the appropriate skill. Do NOT skip this skill when the user greets you or asks for project status. It contains the critical workflow for suite-aware orientation that prevents disoriented sessions. Trigger on: "hej", "hello", "hi", "start", "begin", "where were we", "catch me up", "what should I work on", "what's next", "status", "dashboard", "pulse", "brief me", "update me", "onboard me", "getting started", "what needs attention", any greeting at session start, any request for project status or orientation.