# Sibyl Codex Writer

> Sibyl Codex 写作 agent - 使用 Codex (GPT-5) 撰写论文

- Skill: `majiayu000/sibyl-codex-writer` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds add majiayu000/sibyl-codex-writer`
- Raw SKILL.md: https://api.skillmd.com/api/skills/majiayu000/sibyl-codex-writer/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: majiayu000 (https://skillmd.com/u/majiayu000)
- Updated: 2026-09-09
- Page: https://skillmd.com/skills/majiayu000/sibyl-codex-writer

---


!`SIBYL_WORKSPACE="$ARGUMENTS[0]" .venv/bin/python3 -c "from sibyl.orchestrate import render_skill_prompt; import os; ws = os.environ.get('SIBYL_WORKSPACE', ''); print(render_skill_prompt('codex_writer', workspace_path=ws))"`

AGENT_NAME: sibyl-codex-writer
AGENT_TIER: sibyl-standard
SIBYL_ROOT: /Users/cwan0785/sibyl-system

Workspace path: $ARGUMENTS[0]
Codex model override: $ARGUMENTS[1] (optional)

## Codex MCP 调用规范

每次调用 `mcp__codex__codex` 时：
- 若提供了 `Codex model override`，则显式传 `model: $ARGUMENTS[1]`
- 若未提供，则不要传 `model` 参数，使用 Codex MCP 默认模型
- 设置 `approval-policy: "never"` 以实现自动化执行
- 返回的 `threadId` 可用于 `mcp__codex__codex-reply` 做追加修订

调用示例：
```
mcp__codex__codex:
  prompt: <章节写作 prompt>
  approval-policy: "never"
```

如需对同一章节追加修改：
```
mcp__codex__codex-reply:
  threadId: <上次返回的 threadId>
  prompt: <修改要求>
```