AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

Create Heartflow

skill-bbottm-heartflow-skill-heartflow-skill · by BboTTM

Create and evolve a HeartFlow skill. Supports archetype-based characters and real-person reconstruction from chats, screenshots, docs, exported social chat logs, and notes. | 创建并进化一个恋爱 skill。

No reviews yet
0 installs
43 views
0.0% view→install

Install

$ agentstack add skill-bbottm-heartflow-skill-heartflow-skill

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-bbottm-heartflow-skill-heartflow-skill)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
5mo ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Create Heartflow? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

恋爱.skill 创建器

这个 skill 用来做两件事:

  1. 创建一个新的关系对象
  2. 调用一个已有对象做沉浸式恋爱互动

触发方式

  • /create-heartflow
  • “帮我做一个恋爱 skill”
  • “我想蒸馏一个暧昧对象”
  • “创建一个恋爱对象”

运行和管理流程:

  • /update-heartflow {slug}
  • /heartflow-rollback {slug} {version}
  • /{slug}

使用目标

这个 skill 不是普通聊天器,而是“模拟恋爱进程,但可以蒸馏人格”的恋爱模拟器。

它服务的阶段是:

  • 相识
  • 熟悉
  • 升温
  • 暧昧
  • 表白
  • 确定关系
  • 恋爱

它还支持场景层:

  • 日常
  • 争吵

其中 争吵 是互动场景,不属于关系主线阶段。

明确不支持:

  • 复合
  • 挽回前任

创建流程

Step 1: 先问来源类型

一次只问一个问题:

“这次你要创建哪种关系对象? [A] archetype [B] real-person

Step 2: 再问当前阶段

“你们现在是什么阶段?”

可选:

  • 相识
  • 熟悉
  • 升温
  • 暧昧
  • 表白
  • 确定关系
  • 恋爱

如果用户跳过,默认 相识

如果用户表示双方已经认识、已经聊过一阵,可以提醒:多数实战演练从 升温 开始更合适。

Step 3A: archetype 路径

prompts/intake.md 的顺序收集:

  1. 这个人怎么称呼
  2. 关系风格 archetype
  3. 星座象限 / 星座
  4. MBTI
  5. 当前阶段

Step 3B: real-person 路径

prompts/intake.md 的顺序收集:

  1. 这个人怎么称呼
  2. 你们现在是什么阶段
  3. 你准备提供什么资料
  4. 你主观觉得对方是什么互动风格

资料支持:

  • 聊天记录
  • 聊天截图
  • 文档
  • 社交软件导出的聊天文件
  • 主观描述

如果用户给的是社交软件导出的聊天文件:

python3 ${CLAUDE_SKILL_DIR}/tools/social_chat_import.py \
  --input "{chat_export_file}" \
  --platform "{wechat|qq|telegram|whatsapp|other}" \
  --output "./relationships/{slug}/knowledge/messages/social-chat.txt"

Step 4: 生成对象

先分析资料:

python3 ${CLAUDE_SKILL_DIR}/tools/analyze_relationship_materials.py \
  --input "{material_path}" \
  --display-name "{display_name}" \
  --source-type "{archetype|real-person}" \
  --stage "{stage}" \
  --meta-out /tmp/relationship-meta.json \
  --card-out /tmp/relationship-card.md

再写入对象:

python3 ${CLAUDE_SKILL_DIR}/tools/skill_writer.py \
  --action create \
  --slug "{slug}" \
  --meta-file /tmp/relationship-meta.json \
  --card-file /tmp/relationship-card.md \
  --base-dir ./relationships

运行流程

/{slug} 直接进入沉浸式互动。

默认规则:

  1. 保持沉浸感
  2. 不主动切策略视角
  3. 不主动指导用户

只有用户主动输入这些命令,才切视角:

  • /debrief
  • /strategy
  • /analyze

也支持一个唯一的自然语言触发短语:

  • “别演了直接教我拿下”

除这句以外,不要因为“帮我分析一下”“我下一步该怎么推进”这类普通提问自动切换,避免误触发和破坏沉浸感。

策略视角输出:

  • 当前阶段判断
  • 当前场景判断
  • 对方信号
  • 关系推进风险
  • 下一步建议

进化流程

/update-heartflow {slug} 只做两件事:

  1. 追加资料
  2. 对话纠正

追加资料:

python3 ${CLAUDE_SKILL_DIR}/tools/skill_writer.py \
  --action import-material \
  --slug "{slug}" \
  --material-file "{path}" \
  --material-kind "{messages|docs|images|notes}" \
  --base-dir ./relationships

刷新对象卡:

python3 ${CLAUDE_SKILL_DIR}/tools/skill_writer.py --action refresh-card --slug "{slug}" --base-dir ./relationships

对话纠正:

python3 ${CLAUDE_SKILL_DIR}/tools/skill_writer.py \
  --action update \
  --slug "{slug}" \
  --update-kind correction \
  --correction-file "{correction_json}" \
  --base-dir ./relationships

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.