devcoffee-agent-skills

Use when the user wants to create, update, extend, or replace a Maximus Loop task plan (plan.json). Triggers on "create a plan", "plan this feature", "generate tasks", "break this down into tasks", "add tasks for", "scope this work", "plan the next phase", "update the plan", or /arc:plan.

Use when the user wants to review Maximus Loop results, analyze run performance, or check execution status. Triggers on "review the run", "analyze maximus results", "check status", "how is the run going", "show me the summary", "what's the progress", or when the user wants post-execution analysis with cost/performance insights.

Use when the user asks to "set up maximus", "configure maximus", "initialize maximus", "create maximus config", "scaffold maximus", "bootstrap maximus", or "maximus init".

Use when the user asks to "validate maximus", "check maximus config", "verify maximus setup", "is my maximus config correct", "is my project ready to run", "lint my config", or before running the engine.

Use when the user asks to "list my notes", "show my notes", "what notes do I have", "jot list", "show notes for [project]", or wants a quick overview of captured notes without starting a full review session.

Use when the user asks to "set up maximus", "configure maximus", "initialize maximus", "create maximus config", "scaffold maximus", "bootstrap maximus", or "maximus init".

Use when creating a new Claude Code plugin - enforces architecture patterns, prevents frontmatter bloat, thin wrapper violations, and missing critical fields

Use when implementing any mid-to-large feature, adding functionality, or building components. Triggers on "implement [feature]", "add [functionality]", "build [component]", or explicit "buzzminson" mention. Do NOT use for trivial tasks, package updates, documentation-only changes, or git operations.

Use when reviewing code quality, running code analysis, or performing autonomous review-fix cycles. Triggers on "review my code", "check code quality", "run maximus", "analyze code", or "/maximus". Supports review-only mode (default) and autonomous fix mode (--yolo).

This skill should be used when the user asks to "create an OpenTUI component", "build a TUI screen", "implement terminal UI", "use @opentui/core", "create renderables", "add OpenTUI animation", "build imperative UI", or mentions OpenTUI framework patterns. Provides comprehensive guidance for building terminal UIs with @opentui/core following established best practices from the Maximus Loop TUI POC.

Use when creating Remotion video components, animations, or compositions. Triggers on "create a Remotion component", "build a video animation", "generate Remotion code", "make a text reveal", or when working on Remotion projects and needing component scaffolding.

This skill should be used when the user is working with Remotion, the React-based video creation framework. Trigger when user mentions "Remotion animations", "Remotion composition", "video in React", "useCurrentFrame", "interpolate animations", "Remotion audio", "Remotion captions", "spring animations in Remotion", or asks questions like "how do I animate in Remotion?", "how to add audio to Remotion?", "how to create video compositions?", "how to use Three.js with Remotion?". Also trigger when user is working on React video projects that use frame-based rendering.

Use when initializing a new Remotion project or adding Remotion to an existing project. Triggers on "set up Remotion", "create a new Remotion project", "initialize Remotion", "add Remotion to my app", or when needing Remotion project configuration.

Use when user wants to analyze video files for quality, UI/UX, aesthetics, or technical execution feedback using FFmpeg frame extraction and Claude vision

This skill should be used when the user asks to "jot this down", "note this", "capture a note", "remember this for later", "log a reminder", "save this thought", "make a note", or wants to quickly record a bug, idea, reminder, or improvement for any project.

This skill should be used when the user asks to "review my notes", "triage my jot notes", "go through my notes", "check my notes", "what notes do I have", "review notes for [project]", "/jot:review", or wants to go through their pending jot notes.

Use when releasing a new plugin version - prevents forgotten tags, version mismatches, dirty commits, and minimal CHANGELOGs under time/authority/exhaustion pressure

This skill should be used when the user asks to "check OTA updates", "list updates", "publish an update", "rollback update", "check builds", "list builds", "start a build", "cancel build", "submit to app store", "submit to play store", "manage channels", "create channel", "pause channel", "manage branches", "create branch", "EAS CLI", "npx eas", "check build status", "view update details", "channel rollout", "roll back to embedded", "manage credentials", "set environment variables", "create webhook", "compare fingerprints", "run workflow", "deploy web app", "eas deploy", or any task involving Expo Application Services CLI operations.

Validate a Maximus Loop project configuration. Run deterministic CLI checks and provide project-aware advisories. Use when the user asks to "validate maximus", "check maximus config", "verify maximus setup", "is my maximus config correct", "is my project ready to run", "lint my config", or before running the engine.

This skill should be used when the user asks to "rate this tldr", "evaluate this tldr", "score this summary", "grade this tldr", "how good was that summary", "tldr feedback", or "/tldr:feedback". Handles the complete TLDR evaluation workflow: scoring samples on 4 criteria, creating sample files, and updating the evaluation log. Supports --no-user-score flag for agent-only evaluation when called programmatically by /tldr:note --ex.

This skill should be used when the user asks to "note a tldr improvement", "log a tldr issue", "capture a tldr idea", "jot down a tldr problem", "/tldr:note", or wants to quickly record a bug, feature request, or improvement idea for the TLDR plugin without disrupting their current workflow. Supports an optional --ex flag to include an example of the issue, which auto-triggers evaluation via /tldr:feedback.

This skill should be used when the user asks to "review tldr notes", "triage tldr improvements", "go through tldr backlog", "check my tldr notes", "what tldr notes do I have", "/tldr:review", or wants a dedicated session to discuss, refine, and act on pending improvement notes captured via /tldr:note. Walks through each pending item for discussion and implementation.

This skill should be used when the user asks for a TLDR, summary, recap, or condensed version of content. Invoked as "/tldr" with no arguments to summarize the previous Claude message from conversation context, or as "/tldr [topic]" to summarize whatever the user specifies. Handles requests like "summarize that", "give me the tldr", "recap what you just said", "shorten your last response", or "tldr of the error analysis".

Use when the user wants to create, update, extend, or replace a Maximus Loop task plan (plan.json). Triggers on "create a plan", "plan this feature", "generate tasks", "break this down into tasks", "add tasks for", "scope this work", "plan the next phase", "update the plan", or /maximus-plan. Not for running the engine or monitoring — only for plan creation and modification.

This skill should be used when the user wants to review Maximus Loop results, analyze run performance, or check execution status. Triggers on "review the run", "analyze maximus results", "check status", "how is the run going", "show me the summary", "what's the progress", "/maximus-review", or when the user wants post-execution analysis with cost/performance insights.