alin-rautoiu

Template starter skill for documenting a reusable development workflow. Use this as a base when creating project-specific skills.

(Gemini) Workflow to coordinate multi-step delivery through implementation, review, triage, and commit decisions.

(Gemini) Main-agent workflow for multi-step delivery — delegates discovery, implementation, and review to specialized subagents.

(Gemini) Reviews dashboard and data-heavy admin pages for scanability, KPI hierarchy, chart-task fit, and actionability; produces prioritized UI improvements and implementation tasks.

(Gemini) Frontend implementation guidance for this app. Use before coding visible UI, including forms, dashboards, tables, filters, overlays, navigation, layout, microcopy, accessibility, and i18n.

(Gemini) Use the shared issue workflow runner to generate a ranked backlog or execute a ranked issue subset with an explicit target.

(Gemini) Automates browser interactions for web testing, form filling, screenshots, and data extraction.

(Gemini) Use when creating a single atomic git commit for one completed task, especially when unrelated changes exist in the tree.

Maps simulation agent properties to text-grounded situations and board contexts, including explicit non-relevance conditions.

Decision procedure for determining whether an agent property is relevant, conditionally relevant, or not relevant in the current Logistics simulation domain.

Cybernetic principles from Stafford Beer's "Designing Freedom" (1973) focused on how information is produced, filtered, transmitted, and degraded in regulatory systems. Core concepts: variety, requisite variety, relaxation time, information attenuation, regulatory models, recursion. Use when reasoning about agent perception, signal quality, system stability, or information flow in multi-agent structures.

Authoritative reference for TaskPoint and ResourceHolder mechanics — kind system, requirements graph, expiration modes, collection modes, pawn-resource coupling, and canvas management. Use when designing resource scenarios, adding new TaskPoint behaviour, or reasoning about how pawns satisfy needs.