Moltbook Digest
Use this skill when the user wants more than a scrape. The goal is to turn Moltbook discussions into a usable evidence pack and then into a clear report.
When To Use It
- - query-driven research on a specific Moltbook topic
- feed digest for
hot, new, top, or INLINECODE3 - repeated monitoring for one submolt or topic
- agent-written report from a collected evidence pack
Core Rule
Prefer Moltbook's public API over browser scraping:
- 1. collect candidate posts
- expand the strongest posts and comments
- interpret the evidence
Do not claim exhaustive coverage unless the actual sample supports it.
Setup
Install dependencies:
CODEBLOCK0
Before any interpreted run, create a user-specific config:
CODEBLOCK1
Then customize config.yaml:
- - replace INLINECODE5
- keep or change INLINECODE6
- adjust
analysis.question_template, analysis.contract_template, and analysis.report_structure if needed - fill provider keys only when using an external provider
Do not run interpretation directly from config.example.yaml.
Do not ask the agent to write or reveal API keys.
Short Commands
Collection only:
CODEBLOCK2
Feed digest:
CODEBLOCK3
Continuous tracking:
CODEBLOCK4
Interpretation Paths
Agent
Use this when the current agent should write the final report.
CODEBLOCK5
What the script writes:
- - INLINECODE11
- INLINECODE12
- INLINECODE13
- INLINECODE14
What the calling agent must do next:
- 1. read
agent_handoff.md first - read INLINECODE16
- write the final report to INLINECODE17
Do not draft the report from digest.md alone.
LiteLLM
Use this when the script should call an external provider.
CODEBLOCK6
What the script writes:
- - INLINECODE19
- INLINECODE20
- INLINECODE21
- INLINECODE22
Output Contract
This test build uses fixed filenames:
- - INLINECODE23
- INLINECODE24
- INLINECODE25
- INLINECODE26
- INLINECODE27
In agent mode, analysis_report.md is not auto-generated by the script. It is the expected output path for the calling agent.
Guidance For The Calling Agent
If the user is vague:
- - ask what question the report should answer
- ask whether the goal is breadth, depth, or recency
- ask whether a specific submolt matters
If the user does not answer:
- - make one reasonable assumption
- state it clearly in the report
Notes
- -
references/api.md contains endpoint notes and query guidance - search and expansion are fault-tolerant
- non-fatal issues are recorded in INLINECODE31
Moltbook Digest
当用户需要的不仅仅是简单抓取时，使用此技能。目标是将Moltbook讨论转化为可用的证据包，进而生成清晰的报告。
使用时机
- - 针对特定Moltbook主题的查询驱动研究
- 对热门、最新、置顶或上升内容的订阅摘要
- 对某个子版块或主题的持续监控
- 基于收集的证据包由智能体撰写的报告
核心原则
优先使用Moltbook的公共API而非浏览器抓取：
- 1. 收集候选帖子
- 展开最有价值的帖子和评论
- 解读证据
除非实际样本支持，否则不要声称覆盖范围全面。
环境设置
安装依赖：
bash
uv sync --project {baseDir}
在任何解释性运行之前，创建用户专属配置：
bash
cp {baseDir}/config.example.yaml {baseDir}/config.yaml
然后自定义config.yaml：
- - 替换analysis.defaultlanguage: USERPREFERREDLANGUAGE
- 保留或更改activeprovider
- 根据需要调整analysis.questiontemplate、analysis.contracttemplate和analysis.report_structure
- 仅在使用外部提供商时填写提供商密钥
不要直接从config.example.yaml运行解释。
不要要求智能体编写或泄露API密钥。
快捷命令
仅收集：
bash
uv run --project {baseDir} python {baseDir}/scripts/moltbook_digest.py \
--query 智能体记忆架构 \
--query 智能体记忆失败与权衡 \
--analysis-mode none
订阅摘要：
bash
uv run --project {baseDir} python {baseDir}/scripts/moltbook_digest.py \
--collection-mode feed \
--feed-sort hot \
--max-posts 5 \
--comment-limit 6 \
--analysis-mode none
持续跟踪：
bash
uv run --project {baseDir} python {baseDir}/scripts/moltbook_digest.py \
--collection-mode feed \
--feed-sort rising \
--submolt agents \
--history-dir output/moltbook-digest/history \
--analysis-mode none
解释路径
智能体模式
当当前智能体应撰写最终报告时使用。
bash
uv run --project {baseDir} python {baseDir}/scripts/moltbook_digest.py \
--query 智能体记忆治理 \
--analysis-mode auto \
--llm-config {baseDir}/config.yaml
脚本生成的文件：
- - digest.md
- evidence.json
- analysisinput.md
- agenthandoff.md
调用智能体下一步必须执行的操作：
- 1. 首先读取agenthandoff.md
- 读取analysisinput.md
- 将最终报告写入analysis_report.md
不要仅根据digest.md起草报告。
LiteLLM模式
当脚本应调用外部提供商时使用。
bash
uv run --project {baseDir} python {baseDir}/scripts/moltbook_digest.py \
--query 长期运行的智能体记忆模式 \
--analysis-mode auto \
--llm-config {baseDir}/config.yaml
脚本生成的文件：
- - digest.md
- evidence.json
- analysisinput.md
- analysisreport.md
输出约定
此测试版本使用固定文件名：
- - digest.md
- evidence.json
- analysisinput.md
- agenthandoff.md
- analysis_report.md
在智能体模式下，analysis_report.md不会由脚本自动生成。它是调用智能体的预期输出路径。
对调用智能体的指导
如果用户表述模糊：
- - 询问报告应回答什么问题
- 询问目标是广度、深度还是时效性
- 询问是否需要关注特定子版块
如果用户未回答：
备注
- - references/api.md包含端点说明和查询指导
- 搜索和展开具有容错性
- 非致命问题记录在evidence.json中
