Menu
Skills in this repository
SkillsMP has collected 77 skills from htekdev/copilot-home-assistant. Open a skill to review its source and details.
htekdev/copilot-home-assistantShowing 37 of 77 collected skills.
Agent/skill architecture management for a family assistant platform — find embedded procedures inside agents, extract reusable skills, detect contradictions between skills and agents, and refactor agents to reference skills safely. Use when user says "extract…
Background agent steering protocol — when to use write_agent vs launching fresh, run isolation guards, and cron exception rules. Use when user says "steer agent", "write_agent", "background agent", "run isolation", "inject context", "running agent", "agent…
{{GITHUB_USERNAME}} asset cross-referencing for content creation — blog post discovery, GitHub repo linking, prior post checking, and ecosystem interconnection. Use when user says "cross-reference", "find related articles", "link blog post", "related repos",…
Standardized procedure for managing GitHub Issues in your content pipeline repository as content moves through the pipeline. Use when agent says "update issue", "close issue", "schedule post", "publish content", "reschedule post", "content published", "issue…
Data folder ownership map — which agent owns which data directory, cross-domain read/write rules, and shared file governance. Use when user says "who owns this data", "can I write to", "data ownership", "cross-domain write", "folder ownership", "domain…
Multi-lane pipeline execution pattern with parallel/sequential lanes, retry policies, failure modes, context packages, and coherence correction. Use when user says "pipeline", "multi-lane", "parallel execution", "retry policy", "failure handling",…
Event-to-prep-task generation patterns for a family assistant — calendar scanning, leave-by calculation, prep task creation, smart ordering, and clarification-first workflow. Use when user says "prep tasks", "proactive tasks", "generate tasks from calendar",…
Research document lifecycle management — creating, reusing, updating, superseding, and archiving research across the platform. Ensures research is saved as research (NOT specs), discoverable by all agents, and kept current. Use when user says "research this",…
Monthly/weekly budget report generation — spending breakdown, budget vs actual, trends, upcoming bills, and recommendations. Use when user says "budget report", "monthly review", "spending summary", "budget vs actual", "financial report", "where did money…
Dual-calendar availability checking — combines Google Calendar (personal) and WorkIQ (work/Microsoft 365) to determine true schedule and free slots. Use when agent says "check availability", "find free time", "schedule something", "is {{PARENT_1}} free",…
Structured morning/evening briefing format — weather, dual-calendar merge, tasks, emails, meals, bills, maintenance, and proactive task creation. Use when user says "briefing", "morning summary", "daily report", "what's today look like", "day overview",…
Email subject line encoding rules for gmail_send. Use when sending emails, composing subjects, or any gmail_send call.
Family emergency response procedures — parent notification, medical info relay, child safety escalation, and urgent contact protocols. Use when user says "emergency", "urgent medical", "notify other parent", "emergency contact", "911", "hospital", "safety…
Calendar event classification and OOF block generation — keyword-based category detection, sync rules (personal→work), and Outlook block formatting. Use when user says "categorize event", "OOF block", "sync event", "event category", "what type of event",…
Finance task lifecycle management — auto-pay bill cleanup, payment-logged cluster clearing, and bill reminder task creation/cancellation rules. Use when agent says "bill paid", "auto-pay", "payment logged", "clear reminders", "cancel bill task", "finance…
Smart task ordering algorithm for serving tasks — time-locked first, then urgent/high priority, dependencies, location chaining, energy matching, quick-win momentum, and staleness bumps. Use when user says "next task", "what should I do", "task priority",…
Task template design, trigger configuration, dependency DAG creation, and expansion lifecycle. Use when creating task templates, designing trigger conditions, expanding templates, checking scheduled triggers, or any agent says "create template", "task chain",…
Watch list pattern for tracking expected replies and follow-ups — create watches, check resolution, escalate stale items. Use when user says "watch for reply", "waiting on response", "follow up", "watch list", "track reply", "check watches", "pending…
Weekly planning summary format and procedure — week-in-review, upcoming calendar, priority tasks, meal plan, financial snapshot, and home maintenance. Use when user says "weekly summary", "weekly planning", "Sunday planning", "week review", "weekly report",…
Platform improvement governance — what to auto-fix vs what needs approval, detect-fix-report pattern, and improvement cadence rules. Use when user says "auto improve", "platform fix", "should I ask first", "auto implement", "improvement approval", "detect fix…
No-assumptions clarification protocol — detect missing data, create clarification tasks, block dependent work, and resume only when answers arrive. Use when agent says "clarification needed", "missing info", "don't know current state", "need to ask first",…
Spec-first tiered development pipeline — Research, Plan, Implement, Multi-Model Review, Fix. Use when agent says "make a change", "implement feature", "new extension", "architecture change", "add agent", "refactor", "create spec", "code review", "multi-model…
Standardized error recovery and escalation patterns — failure handling, retry logic, human notification, and graceful degradation. Use when agent says "if fails", "escalate", "notify on error", "retry", "graceful degradation", "error handling", "fallback", or…
Parallel multi-model quality review pattern — launch diverse AI reviewers simultaneously, collect findings, deduplicate, and fix. Use when reviewing articles, code, specs, proposals, or any deliverable that needs quality assurance. Trigger phrases include…
Reusable quality review gate pattern with retry logic, failure escalation, and lessons-learned feedback loops. Use when implementing pre-publish quality checks, content validation gates, automated review-and-retry workflows, or any process that needs "check →…
content pillar definitions for your brand, label system, issue templates, and recording schedule — the reference schema for all content pipeline agents. Use when creating content issues, labeling pipeline items, checking pillar alignment, or referencing the…
Issue reconciliation workflow — keeps GitHub Issues in your content repo in sync with actual Zernio/Late post status. Covers fuzzy matching, label updates, close logic, skip rules, and comment templates. Use when user says "reconcile issues", "sync posts to…
Content queue scheduling maintenance — the 5 ordering rules, near-term scan, deep queue scan, bundle cascade timing, collision detection, and bring-forward optimization for multi-platform social media queues. Use when user says "schedule maintenance", "check…
FFmpeg video editing patterns for creator workflows — silence removal, caption burning, aspect ratio conversion, clip extraction, intro/outro concatenation, and media inspection. Use when user says "edit video", "remove silence", "burn captions", "extract…
Late/Zernio multi-platform social media publishing — account IDs, profile configuration, platform-specific content rules, upload workflow, and post creation patterns for a creator brand. Use when user says "publish post", "schedule post", "create post",…
Video production pipeline for creator brands — editing, quality review, intro/outro concat, publishing, and platform distribution. Use when user says "process video", "publish video", "video pipeline", "edit video", "quality review", "burn captions", "concat…
3-layer lesson persistence pattern — when {{PARENT_1}} or {{PARENT_2}} corrects behavior, persist the lesson across store_memory + standing-orders.md + copilot-instructions.md. Use when user says "persist lesson", "save correction", "never repeat", "remember…
Cron job dispatch rules — always launch fresh agents via task tool, never write_agent for scheduled jobs. Use when user says "cron dispatch", "scheduled job", "launch cron", "cron architecture", "fresh agent for cron", "cron rule", "dispatch pattern", or any…
4-tier memory system management for all stateful agents — loading, saving, pruning, promoting, and maintaining memory files. Use when user says "load memory", "save memory", "update working memory", "prune memory", "memory management", "tier system", or any…
Repo change workflow for safe agent-driven development — fast main-branch edits for tiny changes, and the proper worktree-based isolation flow for larger work. Use when user says "make repo changes", "update repo", "proper workflow", "worktree workflow",…
Safe session restart workflow after creating a NEW custom agent so the new agent becomes available to the task tool. Use when user says "new agent created", "restart for new agent", "agent not available in task tool", "refresh task agent list", or needs a…
Centralized task management procedures for all agents — creating, tracking, updating, and completing tasks via the action-tracker extension. Use when managing tasks, creating action items, checking task status, cleaning up stale tasks, or any agent says…