dasheng-daily-phase2
Use when the workflow enters Stage 2 Brief and the current Agent must generate 8-10 candidate topic cards from canonical intake evidence under the Newma…
Skill by leecyno1
Chalk Labs
Need help deploying dasheng-daily-phase2?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/leecyno1-dasheng-daily-phase2
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: productivity, agent-building
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
- Runs as
- Node.js package
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/dasheng-daily-phase2
What's inside
- Skill name
dasheng-daily-phase2- Description
- Use when the workflow enters Stage 2 Brief and the current Agent must generate 8-10 candidate topic cards from canonical intake evidence under the Newma mainline.
- License
- MIT
- Instructions
- 124 lines in SKILL.md · view on GitHub
Before installing
- Ships executable scripts
Files (8)
- SKILL.md
- brief_card_schema.json
- config.json
- index.js
- ai-generation-guide.md · references/
- content-enhance.md · references/
- frameworks.md · references/
- topic-evaluation.md · references/
SKILL.md (first lines)
# dasheng-daily-phase2
## 目标
这是本仓库 Stage 2 的唯一正式 Brief skill。
职责只有两层:
- 代码层:读取 intake、整理证据包、做显性噪音隔离、校验结构、落盘
- Agent 层:当前 Agent 直接阅读证据包并生成最终独立题卡
失败就失败,不再回退规则造题。
## canonical 输入
按这个优先级读取:
1. `brief_input.json`
2. `channel_top10.json`
3. `event_clusters.json`
4. `raw/intake_records.json`
默认来源目录:
- `${DASHENG_OUTPUT_ROOT:-~/Desktop/自媒体创作}/{run_id}/01_采集/`
## canonical 输出
输出目录:
- `${DASHENG_OUTPUT_ROOT:-~/Desktop/自媒体创作}/{run_id}/02_选题/`
标准产物:
1. `02_编辑Brief库.md`
2. `02_研究Brief库.md`
3. `02_编辑Brief_报告.md`
4. `topic_cards.json`
5. `selected_topics.template.json`
6. `selected_topics.json`
7. `brief_manifest.json`
## 当前规则
- 输出是 **8-10 个平铺独立题卡**
- 不再强制母题 / 变体作为业务主语义
- Agent 可以跨渠道、跨事件综合提炼,但不得脱离 intake 证据池
- Agent 生成的选题标题、判断句和核心题卡字段必须使用中文主体表达;英文来源要先提炼成中文选题
- Brief 记录来源材料大致内容、争议点和已见观点,供 draft 讨论热点,不在本阶段生成完整文章大纲
- Brief 额外沉淀 `question_units` / `opinion_units` / `case_units` / `solution_units`,供 draft 直接取用问题、观点、案例和证明方案,防止只按标题发挥
- 高优先证据池按 `信号强度 × 逻辑独立性 × 主题新颖度` 做弱重排
- intake/hotspot 的 `radar.macro_policy_score` 与 `radar.source_role` 会作为排序元数据进入 Brief,但不得作为硬过滤器
- 不做硬题材配额,但会对同一逻辑链和高重叠签名做边际递减
- Agent 返回的 `existing_evidence` 会回贴到 canonical evidence,并自动补齐 `logic_chain_id`
- 如果同一逻辑链占比超过半数,阶段直接失败
- 默认不调用项目内置 provider;provider 仅作为显式 `--mode provider` 的 legacy/回滚路径
## 执行入口
### 1)准备 Agent 证据包
```bash
python3 scripts/run_mainline_stage.py brief --run-id <run_id>
```
该命令默认生成:
- `brief_signal_bundle.json`
- `brief_agent_prompt.md`
- `brief_manifest.json`(`status=pending_agent_generation`)
### 2)当前 Agent 生成题卡
阅读 `brief_agent_prompt.md` 与 `brief_signal_bundle.json`,生成:
- `topic_cards.agent.json`
根键必须是 `topic_cards`。
### 3)校验并落盘正式 Brief
```bash
python3 scripts/run_mainline_stage.py brief \
--run-id <run_id> \
--agent-cards-file "~/Desktop/自媒体创作/<run_id>/02_选题/topic_cards.agent.json"
```
### 带手动指定题
```bash
python3 scripts/run_mainline_stage.py brief \
--run-id <run_id> \
--manual-topic "你的人工题目"
```
## 参考资料
- Prompt 规则:`../../引擎/03_全链路SOP工作流/02_Brief_AI生成规则.md`
- AI 选题指南:`references/ai-generation-guide.md`
- 选题评估:`references/topic-evaluation.md`
- 写作框架:`references/frameworks.md`
- 内容增强:`references/content-enhance.md`
## 实现说明
- 主脚本:`../../scripts/phase2_rebuilder.py`
- 主链入口:`python3 ../../scripts/run_mainline_stage.py brief --run-id <run_id>`
- Agent finalize:`python3 ../../scripts/run_mainline_stage.py brief --run-id <run_id> --agent-cards-file <file>`
- Schema:`brief_card_schema.json`
- 正式阶段状态:以 `brief_manifest.json` 为准
## 迁移说明
导出仓 `dasheng-stage-brief-ai` 的文档能力已被吸收到这里。
以后不要再把:
- `dasheng-stage-brief-ai`
- `dasheng-daily-brief`From leecyno1/newma-media-studio, MIT.
More from leecyno1/newma-media-studio
- 0.584score
dasheng-account-illustrations
leecyno1
Use when a WeChat article or video needs account-bound concept illustrations.
SkillBFreshness: B★ 5Claude Code +9 - 0.584score
dasheng-sop-orchestrator
leecyno1
Use when running, resuming, or auditing the full Dasheng 8-stage media workflow with strict stage contracts and delivery outputs.
SkillBFreshness: B★ 5Claude Code +9 - 0.584score
dasheng-stage-distribute
leecyno1
Use when entering the distribute stage to publish rewritten content and videos across WeChat, Xiaohongshu, Douyin, Weibo, and X platforms using existing…
SkillBFreshness: B★ 5Claude Code +9 - 0.584score
dasheng-stage-publish-video
leecyno1
Use when entering publish stage and needing the mandatory pre-publish video supplement, including interactive chart videos and motion narrative videos with…
SkillBFreshness: B★ 5Claude Code +9 - 0.584score
dasheng-stage-rewrite
leecyno1
Use when generating the 4-variant rewrite set per topic with hard word-count validation, final-structure inheritance, and optional personal style DNA injection.
SkillBFreshness: B★ 5Claude Code +9 - 0.584score
dasheng-caption-motion
leecyno1
Use when SRT, VTT, ASR segments, or word timestamps must drive animated subtitles, keyword emphasis, kinetic typography, lower thirds, or timed chart callouts…
SkillBFreshness: B★ 5Claude Code +9 - 0.584score
dasheng-commercial-promo-video
leecyno1
Build brand films, product promos, launch trailers, and performance ads as the commercial_promo_video lane.
SkillBFreshness: B★ 5Claude Code +9 - 0.580score
video-download
leecyno1
Download videos from 1800+ websites and generate subtitles using Faster Whisper AI.
SkillBFreshness: B★ 5Claude Code +9 - 0.579score
bilibili-upload-bridge
leecyno1
Use when Newma Publish needs to prepare, upload, or package Bilibili video submissions from completed transwrite video channel packs.
SkillBFreshness: B★ 5Claude Code +9 - 0.579score
dasheng-daily-draft
leecyno1
Use when running Stage 3 Draft from approved Brief topics and producing Reasoning Sheets, standard drafts, and draft quality gates.
SkillBFreshness: B★ 5Claude Code +9 - 0.579score
dasheng-daily-intake
leecyno1
Use when running the canonical Newma intake stage for the current day and producing real-title, real-link source radar outputs for Brief handoff.
SkillBFreshness: B★ 5Claude Code +9 - 0.579score
dasheng-stage-intake-brief-draft
leecyno1
Use when executing the upstream stages (intake, brief, draft) of Dasheng workflow and producing editor-ready topic sets and standard drafts.
SkillBFreshness: B★ 5Claude Code +9
Signals
- Stars
- 5
- Forks
- 1
- Open issues
- 0
- Last commit (this folder)
- Aug 22, 2026 (2 months ago)
- Last release
- Aug 3, 2026 (3 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/leecyno1/newma-media-studio
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from leecyno1
- 0.587score
agent-reach
leecyno1
MUST USE when user wants to 调研/research/搜索/search/查/找/look up anything on the internet — e.g.
SkillBFreshness: B★ 6Claude Code +9 - 0.584score
dasheng-account-illustrations
leecyno1
Use when a WeChat article or video needs account-bound concept illustrations.
SkillBFreshness: B★ 5Claude Code +9 - 0.584score
dasheng-sop-orchestrator
leecyno1
Use when running, resuming, or auditing the full Dasheng 8-stage media workflow with strict stage contracts and delivery outputs.
SkillBFreshness: B★ 5Claude Code +9 - 0.584score
dasheng-stage-distribute
leecyno1
Use when entering the distribute stage to publish rewritten content and videos across WeChat, Xiaohongshu, Douyin, Weibo, and X platforms using existing…
SkillBFreshness: B★ 5Claude Code +9 - 0.584score
dasheng-stage-publish-video
leecyno1
Use when entering publish stage and needing the mandatory pre-publish video supplement, including interactive chart videos and motion narrative videos with…
SkillBFreshness: B★ 5Claude Code +9 - 0.584score
dasheng-stage-rewrite
leecyno1
Use when generating the 4-variant rewrite set per topic with hard word-count validation, final-structure inheritance, and optional personal style DNA injection.
SkillBFreshness: B★ 5Claude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get dasheng-daily-phase2 running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
dasheng-daily-phase2 skill. Agent Mart. https://agentmart.dev/skills/leecyno1-dasheng-daily-phase2 (last updated 2026-10-24)