ワンクリックでManusで任意のスキルを実行

rr

スター2

フォーク0

更新日2026年5月14日 11:41

Research-and-Report protocol for Codex. Invoked by `RR`, `RRP`, `research and report`, broad competitor/market/persona/regulatory research, or explicit research fan-out requests.

インストール

Codex または Claude でインストールこの Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。

Manusで実行

ソース

mreza0100

mreza0100/professor

GitHub リポジトリを開く Creator のリポジトリを見る

ダウンロード

Manusで実行

Codex-only execution mapping

Claude Agent calls map to Codex spawn_agent calls.
Use a scout agent unless the subquestions are already obvious from the request.
For fan-out, spawn 2-6 Codex research agents in parallel, one self-contained RRP per subquestion.
Fan-out agents return findings in chat. They do not write files.
The orchestrator writes exactly one aggregate report under docs/dev/research/ using the RR filename convention.
Do not execute explicit RR in the main conversation with inline WebSearch/WebFetch/grep. Inline search is allowed only for narrow fact checks outside RR.
If the live-agent ceiling prevents all workers from launching at once, fill available slots, wait/close completed agents, and continue the queue until every subquestion is covered.
If Codex child-agent spawning is unavailable, do not pretend inline research was RR. Produce an RRP prompt for the user to run elsewhere, or report the RR execution as blocked by missing agent support.

このリポジトリの他の Skills

同じリポジトリ

wave-build

mreza0100/professor

Cross-project build pipeline. Invoked as $wave-build <feature> or $wave-build with Pipeline/Task manifest/Wave args. Runs the full pipeline end to end including all git work.

2026-07-052

wave

mreza0100/professor

Wave task runner. Reads a wave.md file and runs each pipeline sequentially. Invoked as `$wave <path/to/wave.md>`. Argument is the path to the wave manifest file.

2026-07-052

p-tokens

mreza0100/professor

Attributes Claude Code runtime token spend per sub-agent and per workflow run, parsed from local JSONL transcripts, and ranks the heaviest burners with estimated USD cost. Answers 'which agent burned the most tokens', 'what did this workflow/wave/pipeline cost', 'token breakdown', 'per-operation tokens', and any retrospective spend analysis. Triggered by 'token ledger', 'token attribution', 'heaviest token burner', 'which agent burned the most tokens', 'what did the run/workflow/wave cost', 'token breakdown', 'per-agent tokens', 'per-operation tokens'. Complements /pcm:context-meter: that audits STATIC context size, this covers RUNTIME spend attribution — route static-budget questions to context-meter, after-the-fact spend questions here.

2026-07-052

audit-code-hygiene

mreza0100/professor

Code hygiene audit — ghost fields, dead code, deps, arch, types, naming, quality. Domain-hydrated at install time.

2026-06-172

audit-security

mreza0100/professor

Security deep scan — injection, auth, API, LLM/prompt, sensitive data, crypto, transport, supply-chain. Domain-hydrated at install time.

2026-06-172

p-360

mreza0100/professor

Exhaustive multi-angle analysis. Invoked by `360`, `three-sixty`, `do a 360`, "what could go wrong", or explicit blind-spot sweep requests.

2026-06-172

name	rr
description	Research-and-Report protocol for Codex. Invoked by `RR`, `RRP`, `research and report`, broad competitor/market/persona/regulatory research, or explicit research fan-out requests.

Read .claude/skills/rr/SKILL.md in full — it is the source-of-truth protocol. Follow it verbatim.

Codex-only execution mapping

Claude Agent calls map to Codex spawn_agent calls.
Use a scout agent unless the subquestions are already obvious from the request.
For fan-out, spawn 2-6 Codex research agents in parallel, one self-contained RRP per subquestion.
Fan-out agents return findings in chat. They do not write files.
The orchestrator writes exactly one aggregate report under docs/dev/research/ using the RR filename convention.
Do not execute explicit RR in the main conversation with inline WebSearch/WebFetch/grep. Inline search is allowed only for narrow fact checks outside RR.
If the live-agent ceiling prevents all workers from launching at once, fill available slots, wait/close completed agents, and continue the queue until every subquestion is covered.
If Codex child-agent spawning is unavailable, do not pretend inline research was RR. Produce an RRP prompt for the user to run elsewhere, or report the RR execution as blocked by missing agent support.