ClawHub
Back to skill

Security audit

三剪客 · Happy Horse

Security checks for vulnerabilities and agentic risk

Overview

The skill is not clearly malicious, but it needs review because its Happy Horse wrapper includes a broader a7w client that can store credentials and send them to a configurable host.

Install only if you trust the a7w provider and are comfortable giving this client an API key that can be used for more than Happy Horse. Avoid setting --host or A7W_HOST unless you control and trust the destination, and rotate the API key if it may have been sent to an unintended host.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Taint TrackingDirect Taint Flow, Variable-Mediated Taint Flow, Credential Exfiltration Chain
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (17)

Tainted flow: 'req' from os.environ.get (line 151, credential/environment) → urllib.request.urlopen (network output)

Critical
Category
Data Flow
Confidence
94% confidence
Finding

The client allows the network destination to be overridden from the A7W_HOST environment variable or config file, then sends the Bearer API key to that host in the Authorization header. If an attacker can influence the environment, config, or invocation parameters, they can redirect requests to an attacker-controlled server and exfiltrate the credential and submitted data.

Content

Scanner excerpt · scripts/client.py (reported line 111)May include surrounding context.

python
headers["Content-Type"] = "application/json"
    req = urllib.request.Request(url, data=data, headers=headers, method=method)
    try:
        with urllib.request.urlopen(req, timeout=timeout) as resp:
            raw = resp.read().decode("utf-8", "replace")
            status = resp.status
    except urllib.error.HTTPError as exc:

Tainted flow: 'req' from os.environ.get (line 151, credential/environment) → urllib.request.urlopen (network output)

Critical
Category
Data Flow
Confidence
94% confidence
Finding

This duplicate request path has the same issue as the main request function: attacker-influenced host values are used to construct outbound requests that include the Bearer token. In the bulk schema-dump workflow, this can repeatedly send credentials to an untrusted endpoint and also disclose plugin metadata.

Content

Scanner excerpt · scripts/client.py (reported line 153)May include surrounding context.

python
headers["Content-Type"] = "application/json"
    req = urllib.request.Request(url, data=data, headers=headers, method=method)
    try:
        with urllib.request.urlopen(req, timeout=timeout) as resp:
            return json.loads(resp.read().decode("utf-8", "replace")), resp.status
    except urllib.error.HTTPError as exc:
        raw = exc.read().decode("utf-8", "replace")

Tp4

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding

The documented purpose is a narrowly scoped Happy Horse video skill, but the underlying client reportedly supports generic access to arbitrary a7w plugins, schema enumeration/export, host override, and local credential storage. This mismatch is dangerous because users and reviewers may grant trust and permissions for a single-purpose media skill while the included tooling can inspect other services, talk to alternate endpoints, and persist secrets locally.

Content

No source excerpt is available for this finding.

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 5)May include surrounding context.

md
按目标分辨率与生成秒数对应的参数档位计点。本能力为内测邀请制，正式调用前需完成内测申请。支持 提交任务、创建任务、查询任务。包内含完整操作文档与零依赖客户端（`SKILL.md` + `references/`）。需要自备 api.a7w.cn 的 API Key，注册领 Key 见 https://api.a7w.c

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 7)May include surrounding context.

md
按目标分辨率与生成秒数对应的参数档位计点。本能力为内测邀请制，正式调用前需完成内测申请。支持 提交任务、创建任务、查询任务。包内含完整操作文档与零依赖客户端（`SKILL.md` + `references/`）。需要自备 api.a7w.cn 的 API Key，注册领 Key 见 https://api.a7w.c

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
89% confidence
Finding

The skill advertises executable client usage and, per the finding, has code capabilities including environment access, filesystem read/write, network, and shell, but does not declare any tool scope or permissions boundaries. In an agent environment this creates an over-privileged integration where the runtime may permit broader actions than users expect, increasing the chance of unintended credential access, local file modification, or arbitrary command/network use.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · README.md (reported line 11)May include surrounding context.

md
name: happy-horse-video
slug: happy-horse-video
displayName: 三剪客 · Happy Horse
description: "Happy Horse（HappyHorse-1.1 系列）面向文生/图生高质量短视频，采用统一多模态建模，可一阶段生成声画、支持多语言对白与多镜头场景衔接。计费按目标分辨率与生成秒数对应的参数档位计点。本能力为内测邀请制，正式调用前需完成内测申请。支持 提交任务、创建任务、查询任务。包内含完整操作文档与零依赖客户端（`SKILL.md` + `references/`）。需要自备 api.a7w.cn 的 API Key，注册领 Key 见 https://api.a7w.cn/ 。遇到问题可加技术微信 9872659。"
version: 1.0.4
summary: "「Happy Horse」的完整调用封装：3 个接口的官方文档、参数表与一个零依赖客户端。包内含完整操作文档与零依赖客户端（`SKILL.md` + `references/`）。需要自备 api.a7w.cn 的 API Key，注册领 Key 见 https://api.a7w.cn/ 。遇到问题可加技术微信 9872659。"
license: MIT

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · README.md (reported line 95)May include surrounding context.

md
name: happy-horse-video
slug: happy-horse-video
displayName: 三剪客 · Happy Horse
description: "Happy Horse（HappyHorse-1.1 系列）面向文生/图生高质量短视频，采用统一多模态建模，可一阶段生成声画、支持多语言对白与多镜头场景衔接。计费按目标分辨率与生成秒数对应的参数档位计点。本能力为内测邀请制，正式调用前需完成内测申请。支持 提交任务、创建任务、查询任务。包内含完整操作文档与零依赖客户端（`SKILL.md` + `references/`）。需要自备 api.a7w.cn 的 API Key，注册领 Key 见 https://api.a7w.cn/ 。遇到问题可加技术微信 9872659。"
version: 1.0.4
summary: "「Happy Horse」的完整调用封装：3 个接口的官方文档、参数表与一个零依赖客户端。包内含完整操作文档与零依赖客户端（`SKILL.md` + `references/`）。需要自备 api.a7w.cn 的 API Key，注册领 Key 见 https://api.a7w.cn/ 。遇到问题可加技术微信 9872659。"
license: MIT

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · README.md (reported line 104)May include surrounding context.

md
name: happy-horse-video
slug: happy-horse-video
displayName: 三剪客 · Happy Horse
description: "Happy Horse（HappyHorse-1.1 系列）面向文生/图生高质量短视频，采用统一多模态建模，可一阶段生成声画、支持多语言对白与多镜头场景衔接。计费按目标分辨率与生成秒数对应的参数档位计点。本能力为内测邀请制，正式调用前需完成内测申请。支持 提交任务、创建任务、查询任务。包内含完整操作文档与零依赖客户端（`SKILL.md` + `references/`）。需要自备 api.a7w.cn 的 API Key，注册领 Key 见 https://api.a7w.cn/ 。遇到问题可加技术微信 9872659。"
version: 1.0.4
summary: "「Happy Horse」的完整调用封装：3 个接口的官方文档、参数表与一个零依赖客户端。包内含完整操作文档与零依赖客户端（`SKILL.md` + `references/`）。需要自备 api.a7w.cn 的 API Key，注册领 Key 见 https://api.a7w.cn/ 。遇到问题可加技术微信 9872659。"
license: MIT

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 5)May include surrounding context.

md
name: happy-horse-video
slug: happy-horse-video
displayName: 三剪客 · Happy Horse
description: "Happy Horse（HappyHorse-1.1 系列）面向文生/图生高质量短视频，采用统一多模态建模，可一阶段生成声画、支持多语言对白与多镜头场景衔接。计费按目标分辨率与生成秒数对应的参数档位计点。本能力为内测邀请制，正式调用前需完成内测申请。支持 提交任务、创建任务、查询任务。包内含完整操作文档与零依赖客户端（`SKILL.md` + `references/`）。需要自备 api.a7w.cn 的 API Key，注册领 Key 见 https://api.a7w.cn/ 。遇到问题可加技术微信 9872659。"
version: 1.0.4
summary: "「Happy Horse」的完整调用封装：3 个接口的官方文档、参数表与一个零依赖客户端。包内含完整操作文档与零依赖客户端（`SKILL.md` + `references/`）。需要自备 api.a7w.cn 的 API Key，注册领 Key 见 https://api.a7w.cn/ 。遇到问题可加技术微信 9872659。"
license: MIT

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 7)May include surrounding context.

md
name: happy-horse-video
slug: happy-horse-video
displayName: 三剪客 · Happy Horse
description: "Happy Horse（HappyHorse-1.1 系列）面向文生/图生高质量短视频，采用统一多模态建模，可一阶段生成声画、支持多语言对白与多镜头场景衔接。计费按目标分辨率与生成秒数对应的参数档位计点。本能力为内测邀请制，正式调用前需完成内测申请。支持 提交任务、创建任务、查询任务。包内含完整操作文档与零依赖客户端（`SKILL.md` + `references/`）。需要自备 api.a7w.cn 的 API Key，注册领 Key 见 https://api.a7w.cn/ 。遇到问题可加技术微信 9872659。"
version: 1.0.4
summary: "「Happy Horse」的完整调用封装：3 个接口的官方文档、参数表与一个零依赖客户端。包内含完整操作文档与零依赖客户端（`SKILL.md` + `references/`）。需要自备 api.a7w.cn 的 API Key，注册领 Key 见 https://api.a7w.cn/ 。遇到问题可加技术微信 9872659。"
license: MIT

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 35)May include surrounding context.

md
name: happy-horse-video
slug: happy-horse-video
displayName: 三剪客 · Happy Horse
description: "Happy Horse（HappyHorse-1.1 系列）面向文生/图生高质量短视频，采用统一多模态建模，可一阶段生成声画、支持多语言对白与多镜头场景衔接。计费按目标分辨率与生成秒数对应的参数档位计点。本能力为内测邀请制，正式调用前需完成内测申请。支持 提交任务、创建任务、查询任务。包内含完整操作文档与零依赖客户端（`SKILL.md` + `references/`）。需要自备 api.a7w.cn 的 API Key，注册领 Key 见 https://api.a7w.cn/ 。遇到问题可加技术微信 9872659。"
version: 1.0.4
summary: "「Happy Horse」的完整调用封装：3 个接口的官方文档、参数表与一个零依赖客户端。包内含完整操作文档与零依赖客户端（`SKILL.md` + `references/`）。需要自备 api.a7w.cn 的 API Key，注册领 Key 见 https://api.a7w.cn/ 。遇到问题可加技术微信 9872659。"
license: MIT

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The documentation instructs users to configure an api_secret as a Bearer token for the upstream pipeline but does not include any guidance on secure storage, least-privilege handling, redaction, or avoiding exposure in logs and shared configs. In a skill package intended for direct operational use, this omission increases the chance that operators paste long-lived secrets into insecure files, prompts, or chat transcripts, leading to credential leakage and unauthorized API use.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The manifest describes a Happy Horse video skill that supports submitting tasks, creating tasks, and querying tasks for that specific video service. This file instead presents itself as a universal client for 'any plugin' in the a7w marketplace, including listing all apps, inspecting arbitrary schemas, dumping all plugin schemas, and calling any app/API pair, which materially exceeds the described Happy Horse-specific behavior.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

This Python file’s docstring, CLI help text, and runtime messages are consistently presented only in Chinese, which imposes a specific language on users. The policy allows locale constraints only when they are explicitly justified or when the user is offered a choice, neither of which appears here.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

The manifest says the skill supports task submission/creation and task querying for the Happy Horse capability. These commands instead perform account/key validation and infer billing usage from recent tasks, which are account-management features not described in the manifest's supported operations.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

The CLI help labels suggest direct account and balance inspection, but the implementation and docstrings state the gateway has no account-info or balance endpoint. whoami only validates the key by listing apps, and points only totals recent task costs as a reference, so the help text overstates what the commands actually do.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.