Agent Building skills

Browse reusable Agent Skills, each with a clear purpose and practical guidance.

seedance-prompts

Write effective prompts for Seedance 2.0 (bytedance/seedance-2.0) multimodal video generation — the official basic prompt formula (Subject+Motion required; Environment/Aesthetics/Camera/Audio optional), the @ reference system (@Image1/@Video1/@Audio1 role assignment, equivalent to the guide's plain Image 1/Video 1), camera-language vocabulary, time-segmented prompt structure, in-video text (slogans/titles, subtitles, speech bubbles), and capability-specific patterns (character consistency, camera/effects replication, video extension, editing, track completion / multi-clip stitching, music beat-matching, dialogue, one-take, e-commerce, educational). A craft overlay on the art-director step: the output prompt is fed to `ralphy generate video --model bytedance/seedance-2.0`, never to a provider API directly. USE WHEN the user is generating video with seedance and needs a prompt drafted or improved, asks "write a seedance prompt", "how do I reference my uploaded image/video in the prompt", wants beat-synced cuts / camera replication / video extension on seedance, or the art-director playbook has already picked seedance for the job (stylized motion, horror/POV/walking, non-default physics). DO NOT FIRE for model SELECTION (that is MODELS.md + the art-director playbook), for Kling/Veo prompt drafting, or for photoreal-human i2v anchors (seedance's compliance filter blocks them — route to kling-v3.0-pro instead).

108 repo starsObserved in 1 repos
Agent Building
GitHub

skills-list

List all Oracle skills with profile tier, type, and script status. Use when user says "skills list", "show skills", "what skills", "list all skills", "how many skills", or wants to see available skills by profile.

108 repo starsObserved in 1 repos
Agent Building
GitHub

tavern-card-distiller

将 SillyTavern / TavernAI 酒馆角色卡（PNG/WEBP/JPEG/JSON 格式）蒸馏为可用的 Claude Code skill，实现沉浸式角色扮演对话。支持 V1/V2/V3 角色卡规范、角色定义、世界书、正则、内置插图和 AI 生成插图。触发词：角色卡、酒馆卡、蒸馏角色、tavern card、character card、导入角色卡、角色卡转 skill、蒸馏角色卡。

108 repo starsObserved in 1 repos
Agent Building
GitHub

team-agents

Spin up coordinated agent teams for any task. Reusable framework for TeamCreate/SendMessage/TaskList patterns. Use when user says "team-agents", "spin up a team", "use teammates", "parallel agents", "coordinate agents", "fan out", or wants multiple agents working together with coordination. Do NOT trigger for simple subagent work (use Agent tool directly) or inter-Oracle messaging (use /talk-to).

108 repo starsObserved in 1 repos
Agent Building
GitHub

universe-studio

Studio mode for a tagged workspace — the one-tag, four-approvals orchestrator (#474). Given an idea and an active workspace (`ralphy workspace use <ws>`), it runs production end-to-end as FOUR stages, each auto-assembled and auto-repaired against that workspace's own eval rubric before you see it: (1) location/cast → phase `style-lock`; (2) scenario → phase `scenario`; (3) scene anchors → phase `assets`; (4) final montage → phase `eval`. Each stage: assemble → `ralphy workspace eval <id>` → read the per-criterion scorecard → the #473 bounded repair loop (FREE editor fixes auto-loop; PAID regeneration STOPS for your approval) → present ONLY when the stage's criteria clear → wait for your stage approval before advancing. You make four decisions; everything else repairs itself against the universe's hard bar. GENERIC: parameterized by the ACTIVE workspace — works for ANY workspace with an `evaluators.json` + `stageGates`. It SEQUENCES the per-role playbooks (scenarist / art-director / editor) under the eval gates; it does NOT replace them. USE WHEN the user says "studio mode", "run my universe end-to-end", "make a <universe> video with the gates", "I'll make four approvals", "auto-assemble and auto-repair against my evals", "drive the staged studio for <workspace>", or tags a workspace and drops an idea expecting the staged-gate flow. See body for ALSO FIRE / DO NOT FIRE / HARD INVARIANTS.

108 repo starsObserved in 1 repos
Agent Building
GitHub

waves-codex

WAVES - Workers, Aggregate, Verify, Extend - wave-based orchestration for Codex. Decompose a big goal into independent slices, verify coverage, spawn Codex subagents in parallel as a bounded wave, collect evidence-backed handoffs, verify important claims, synthesize one deliverable, and extend into another wave only when warranted. Bounded by design to avoid runaway token loops; invoke deliberately. Formerly parallel-orchestrate-codex; also fan out, parallelize, spin up multiple agents, orchestrate workers, multi-stream research, audit a repo, split disjoint implementation work.

108 repo starsObserved in 1 repos
Agent Building
GitHub

aegis-setup

Guide for installing and initializing Aegis MCP server in a project. Use when setting up Aegis, adding architecture enforcement, initializing aegis, or when the user mentions aegis setup, onboarding, or project initialization with Aegis.

107 repo starsObserved in 5 repos
Agent Building
GitHub

initiate-memory

Comprehensive guide for initializing or reorganizing agent memory and project context. Use when setting up a new project, when the user asks you to learn about the codebase, or when you need to create effective memory blocks for project conventions, preferences, and workflows.

107 repo starsObserved in 3 repos
Agent Building
GitHub

system-init

Initialize or hydrate the agent's memory system and verify configuration.

107 repo starsObserved in 3 repos
Agent Building
GitHub