本文へ移動
cccskills
無料GitHub で公開

call-claude

Delegate a prompt to Claude via `claude -p`. Use when the user wants to offload a task to Claude from another agent (Codex, Cursor, another Claude session), get a second opinion, or run a parallel implementation. Default model is opus.

stablyai/call-claude
インストール方法を見る

含まれるファイル（2）

  • SKILL.md1.7 KB
  • call-claude.sh1.1 KB

SKILL.md（原文）

インストールする前に、エージェントに与えられる指示の中身を確認できます。

call-claude

Delegate an arbitrary prompt to Claude and return the output.

Arguments

Required: a prompt.

Optional flags (parsed from the user's message):

FlagValuesDefault
--modelopus, sonnet, haiku, or a full model IDopus
--effortlow, medium, high, xhigh, max(omitted)

Process

  1. Parse the user's message — extract the prompt and any --model / --effort overrides.
  2. Invoke the wrapper script that ships alongside this SKILL.md: 
    "$(dirname "$0")/call-claude.sh" [--model <alias>] [--effort <level>] "<prompt>"
    For multi-line prompts, pipe via stdin: 
    echo "<prompt>" | "$(dirname "$0")/call-claude.sh" --model opus
  3. Return Claude's full output to the user. Don't summarize unless asked.

claude -p is non-interactive and prints the final answer to stdout, so the wrapper output can be captured directly.

Prerequisite

Claude Code must be installed — verify with claude --version.

Examples

# Default (opus)
./call-claude.sh "explain what this repo does"

# Override model
./call-claude.sh --model sonnet "summarize the README"

# Override model + effort
./call-claude.sh --model opus --effort high "design a caching layer"

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

ログインしてレビューを書く
このスキルの問題を報告する