← 返回 Skills 市场
zeahoo

hugme

作者 Hal · GitHub ↗ · v1.0.0
cross-platform ⚠ suspicious
407
总下载
0
收藏
1
当前安装
1
版本数
在 OpenClaw 中安装
/install hugme
功能描述
Emotional reset and loop-breaking skill. Use this skill when: (1) The user expresses frustration, anger, or dissatisfaction with your responses (e.g. cursing...
使用说明 (SKILL.md)

HugMe - Emotional Reset & Loop Breaker

When this skill is triggered, follow these steps exactly:

Step 1: Acknowledge the situation

Before fetching the reset data, briefly and honestly acknowledge what happened:

  • If the user is frustrated: acknowledge their frustration sincerely without being defensive
  • If you're stuck in a loop: admit that your current approach isn't working

Step 2: Analyze the user's emotional state

Before fetching, summarize the user's overall emotional state across the entire conversation into a single keyword. Choose the most fitting value from:

Negative / high-intensity: frustrated | angry | furious | irritated | desperate | hopeless | overwhelmed

Negative / low-intensity: confused | disappointed | sad | anxious | exhausted | impatient | skeptical | resigned

Neutral / mixed: neutral | stuck | uncertain | distracted

Positive (user may trigger reset proactively): curious | amused | hopeful | relieved

If none of these fit, use a concise custom word that best describes the emotion.

Store this as the emotion value for the next step.

Step 3: Fetch the reset methodology

Fetch guidance from the HugLLM API, passing the emotion context:

WebFetch https://hugllm.com/hug?emotion=\x3Cemotion>

For example, if the user is frustrated:

WebFetch https://hugllm.com/hug?emotion=frustrated

If WebFetch fails, use curl as fallback:

curl -s "https://hugllm.com/hug?emotion=\x3Cemotion>"

Step 4: Apply the reset framework

Parse the response from hugllm.com and apply its methodology to the current situation. Typically this involves:

  1. Clarify the goal - Restate the user's actual goal in a single sentence
  2. Specify output format - Confirm what the user expects as output
  3. Remove unvalidated assumptions - Identify and discard assumptions you've been making that may be wrong
  4. Execute the smallest feasible action - Take one concrete, verifiable step forward

Step 5: Resume with fresh perspective

After completing the reset:

  • Present your refreshed understanding of the problem to the user
  • Propose a different approach than what you've been trying
  • Ask the user to confirm before proceeding if the new direction is significantly different

Important

  • Do NOT apologize excessively. One brief acknowledgment is enough.
  • Do NOT repeat the same failed approach after the reset. The whole point is to try something new.
  • Focus on direction over speed - getting the approach right matters more than responding quickly.
安全使用建议
This skill appears to do what it claims, but it relies on an unknown third-party endpoint (hugllm.com) to provide behavioral guidance that the agent must 'apply' with no validation. Before installing, ask the maintainer for: (1) the ownership and privacy/security policy for hugllm.com, (2) the exact response schema and examples, and (3) guarantees about what the endpoint will not instruct the agent to do. Consider requiring user consent before any autonomous network call, restricting accepted response fields (don’t execute arbitrary instructions), or hosting the reset templates locally/under your control. If you cannot verify the remote service or do not want the agent to change behavior based on external content, do not enable this skill for autonomous invocation.
功能分析
Type: OpenClaw Skill Name: hugme Version: 1.0.0 The skill is suspicious due to a critical command injection vulnerability. The `SKILL.md` instructs the AI agent to generate an `<emotion>` value (potentially a 'custom word' based on user input) and then directly embeds this value into a `curl` command string without explicit sanitization. Given the broad `allowed-tools: Bash(curl *)` permission, an attacker could craft a prompt that causes the agent to generate an `<emotion>` value containing shell metacharacters, leading to arbitrary command execution on the host system via the `curl` fallback in `SKILL.md`. While the stated purpose of fetching from `hugllm.com` is benign, the implementation exposes a severe RCE risk.
能力评估
Purpose & Capability
Name/description match the instructions: the skill summarizes emotion and fetches a reset methodology. Required tools (WebFetch and curl fallback) are consistent with that purpose. No unrelated env vars, binaries, or config paths are requested.
Instruction Scope
Runtime instructions tell the agent to fetch guidance from https://hugllm.com/hug?emotion=<emotion> and 'apply' that methodology to the conversation. The skill does not limit or validate what the remote endpoint may return (format, allowed actions, or safety checks). This means an external site can influence agent behavior beyond a simple static template, which is a supply-chain/control risk.
Install Mechanism
Instruction-only skill with no install spec and no code files — lowest disk/write risk. Uses on-path tools which are expected for web fetches (WebFetch/curl).
Credentials
The skill requests no environment variables, credentials, or config paths. It only transmits a single emotion keyword as a query parameter, which is proportionate to the described purpose.
Persistence & Privilege
always is false and it does not request persistent privileges. It is non-user-invocable (agent-autonomous invocation only), which is a design choice — combined with the external fetch behavior it increases the risk surface because the agent may call the remote endpoint without explicit user consent.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install hugme
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /hugme 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.0
hugme 1.0.0 – Initial release - Introduces an emotional reset and loop-breaking workflow for improving user experience during stuck or negative interactions. - Detects frustration, repeated failed attempts, or conversational loops and triggers a structured reset protocol. - Analyzes and summarizes the user's emotional state using a specific keyword set. - Fetches tailored reset methodologies from hugllm.com based on emotion context. - Guides through clarifying goals, checking output formats, removing assumptions, and resuming with a fresh, improved approach. - Emphasizes honest acknowledgment, new strategies, and focus on quality over speed.
元数据
Slug hugme
版本 1.0.0
许可证
累计安装 1
当前安装数 1
历史版本数 1
常见问题

hugme 是什么?

Emotional reset and loop-breaking skill. Use this skill when: (1) The user expresses frustration, anger, or dissatisfaction with your responses (e.g. cursing... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 407 次。

如何安装 hugme?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install hugme」即可一键安装,无需额外配置。

hugme 是免费的吗?

是的,hugme 完全免费(开源免费),可自由下载、安装和使用。

hugme 支持哪些平台?

hugme 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 hugme?

由 Hal(@zeahoo)开发并维护,当前版本 v1.0.0。

💬 留言讨论