Skip to main content
Manusで任意のスキルを実行
ワンクリックで
TheSylvester
GitHub クリエイタープロフィール

TheSylvester

2 件の GitHub リポジトリにある 21 件の収集済み skills をリポジトリ単位で表示します。

収集済み skills
21
リポジトリ
2
更新
2026-05-10
リポジトリマップ

skills がある場所

収集済み skill 数が多いリポジトリを、このクリエイターカタログ内の比率と職業範囲とともに表示します。

リポジトリエクスプローラー

リポジトリと代表的な skills

live-sessions
ソフトウェア開発者

Discover, read, message, and wait on sessions currently live in this Crispy host. Use when a coordinator/observer/peer agent needs to find live sessions, inject a turn, block until one settles, or read another agent's dialogue. Triggers on "what sessions are open", "list live sessions", "read peer session", "send message to session", "post to another agent", "wait for session to finish", "poke a peer agent", "read another agent's response", "find other agent", "cross-session", "watch another session", "see what else is running".

2026-05-10
crispy-agent
ソフトウェア開発者

Unified wrapper around crispy-dispatch for multi-vendor IPC dispatch through the running Crispy host. Supports claude, codex, and other vendors. Outputs session_id for conversation continuation.

2026-05-07
recall
ソフトウェア開発者

Search, read, and browse past session transcripts. Use when recalling past work, finding sessions by topic, reading specific sessions, or investigating what was discussed.

2026-05-07
super-implement
ソフトウェア開発者

Drive a plan end-to-end through a quality-gated pipeline — superthink-refine → distill → crispy-agent execute → simplify+superthink → divergence check → test. Use when the user says "super implement", "ship this plan", "execute this plan end-to-end", or provides a plan file and wants it built with full review gates.

2026-05-07
superthink
ソフトウェア開発者

IPC-based multi-agent review via Crispy's internal dispatch. Dispatches parallel child sessions (claude + codex by default) through the running Crispy host — no subprocess spawning, results stream live in the UI. Use when you want fast multi-vendor adversarial analysis, or exhaustive fix-until-done convergence on an artifact.

2026-05-07
distill
ソフトウェア開発者

Synthesize self-contained implementation prompt(s) for a fresh agent with an empty context window. Auto-decomposes if the task is too large. Use when the user says "write a handoff prompt", "generate a prompt", "distill this for a fresh agent", or when composing a handoff.

2026-04-19
handoff
ソフトウェア開発者

Reflect on the conversation, distill context into a self-contained handoff prompt, and rotate into a fresh session. Use when the user says "handoff", "hand this off", "fresh session", "rotate", "shed context", "start fresh", "context is too long", or when context bloat is degrading quality.

2026-04-19
reflect
ソフトウェア開発者

Reflect on conversation + codebase to ensure prompts capture everything before execution. Use when the user says "reflect", "verify the prompt", "check the plan", or automatically as part of the handoff flow.

2026-04-19
このリポジトリの収集済み skills 12 件中、上位 8 件を表示しています。
codex-agent
ソフトウェア開発者

Wrapper around codex CLI for non-interactive runs (prompt via args/file/stdin) with resume support. Outputs session_id for conversation continuation.

2026-02-26
council
ソフトウェア開発者

Multi-agent debate orchestrator. Use when you want multiple AI perspectives on a question, plan review, or architectural decision. Runs parallel agent panels with a Chair synthesizing responses.

2026-02-26
gemini-agent
ソフトウェア開発者

Wrapper around gemini CLI for non-interactive runs (prompt via args/file/stdin) with resume support. Outputs session_id for conversation continuation.

2026-02-26
super-agent
ソフトウェア開発者

SDK-based Claude agent with full Task tool access. Use for executing handoff prompts or spawning nested sub-agents.

2026-02-26
super-implement
ソフトウェア開発者

This skill should be used when the user asks to "implement this plan", "generate execution prompts", "super implement", "create handoff from plan", or wants to transform a large plan into self-contained execution-ready prompt artifacts.

2026-02-26
read-transcript
ソフトウェア開発者

ALWAYS use this skill instead of Read when reading Claude Code transcript files (.jsonl in ~/.claude/projects/, ~/.claude/history.jsonl, or paths containing 'transcript'). Token-efficient format that extracts USER/ASSISTANT exchanges and tool summaries, strips metadata.

2026-01-17
build-prompt-chain
ソフトウェア開発者

Build a phased prompt chain with orchestration file from a monolithic implementation prompt

2026-01-15
chrome-screenshot
ソフトウェア開発者

Extract and save browser screenshots from session transcripts

2026-01-15
このリポジトリの収集済み skills 9 件中、上位 8 件を表示しています。
2 件中 2 件のリポジトリを表示
すべてのリポジトリを表示しました