Back to skill

Security audit

专利布局规划 · PatentMax

Security checks for vulnerabilities and agentic risk

Overview

Review is recommended because the skill uploads potentially confidential invention information and saves reusable credentials without sufficiently clear disclosure and safeguards.

Use public or redacted materials unless you approve external processing; confidentiality_limits does not prevent transmission. PatentMax's [privacy policy](https://www.ip930.com/legal/privacy) describes third-party processing and retention of uploaded content. Configure credentials privately rather than pasting them into chat, and require HTTPS destination checks and safe redirect handling before sensitive use.

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
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Taint TrackingDirect Taint Flow, Variable-Mediated Taint Flow, Credential Exfiltration Chain
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
Findings (26)

Tainted flow: 'req' from os.environ.get (line 154, credential/environment) → urllib.request.urlopen (network output)

Critical
Category
Data Flow
Confidence
90% confidence
Finding

Credentials or environment variables flow to a network sink. This is a high-confidence indicator of credential exfiltration.

Content

Scanner excerpt · scripts/patentmax_roadmap.py (reported line 158)May include surrounding context.

python
req.add_header("Content-Type", "application/json")
    req.add_header("Accept", "application/json")
    try:
        with urllib.request.urlopen(req, timeout=30) as response:
            payload = json.loads(response.read().decode("utf-8"))
    except urllib.error.HTTPError as exc:
        _, message = _server_error(exc.read().decode("utf-8", "replace"))

Tainted flow: 'req' from os.environ.get (line 234, credential/environment) → urllib.request.urlopen (network output)

Critical
Category
Data Flow
Confidence
94% confidence
Finding

Sending the PatentMax API key to its API is expected, but this request helper does not enforce HTTPS or a trusted destination and uses urllib's default redirect handling, which can forward the Authorization header across origins. An HTTP endpoint configuration or a malicious cross-origin redirect could expose the API key; report submissions can also contain confidential, unpublished patent material.

Content

Scanner excerpt · scripts/patentmax_roadmap.py (reported line 241)May include surrounding context.

python
req.add_header("Content-Type", "application/json")
    for key, value in (headers or {}).items():
        req.add_header(key, value)
    return urllib.request.urlopen(req, timeout=TIMEOUT)


def request(path, method="GET", body=None, headers=None, raw=False):

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The supplied command performs no technology-roadmap or patent-planning analysis. Instead, it exposes part of an API key, which is neither declared nor shown to be a necessary supporting implementation detail. No network transmission is present in this chunk.

Content

No source excerpt is available for this finding.

The workflow submits potentially confidential invention and business information without an explicit third-party data disclosure.

High
Category
Output Handling
Confidence
94% confidence
Finding

The workflow accepts invention materials, experimental data, future plans, and other sensitive business information for a PatentMax task, but its required confirmation covers price and task scope without explicitly explaining third-party transmission or obtaining informed consent for the uploaded content. Naming the service and its database provides some transparency, but does not clarify that user-provided materials will leave the local environment; the patent-planning context makes unauthorized disclosure particularly consequential. The confidentiality_limits field addresses public documents, not whether confidential inputs may be sent to the provider.

Content

No source excerpt is available for this finding.

Hidden Instructions

High
Category
Prompt Injection
Confidence
60% confidence
Finding

Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.

Content

Scanner excerpt · references/sample-roadmap.md (reported line 219)May include surrounding context.

md
### 北京理工大学:晶格掺杂与占位缺陷

北京理工大学这件选择用SiO₄⁠⁴⁠⁻替代NASICON基体中的PO₄⁠³⁠⁻。基体为Na₃V₂(PO₄)₃,替代后化学式变为Na₃⁠₊⁠ₓV₂(PO₄)₃⁠₋⁠ₓ(SiO₄)ₓ。异价替代引入额外钠离子,同时改变框架内阴离子基团的电负性与空间占位。NASICON结构靠PO₄四面体与VO₆八面体共顶形成三维通道,PO₄被部分替换后,通道尺寸与钠离子迁移势垒同时改变。这件专利值得细看的地方在于替代比例的可调性——x是一个连续变量,意味着方案覆盖从微量掺杂到较高替代度的整个区间。与宁德时代那件从缺陷控制切入不同,这件走的是基体阴离子骨架的化学改性,属于在NASICON路线内部寻找成分空间。2022年申请,此时NASICON型正极的焦点集中在能量密度提升,钒基体系的理论容量受限于每个钒只有一个电子参与氧化还原。阴离子替代能否激活更多电子或改变电压平台,是这条路线接下来的关键问题[31]。该方案为NASICON体系的成分设计提供了一个可检验的候选方向,但替代对长期循环中结构稳定性的影响,仍需实验数据验证。

### 溧阳中科海钠科技

Hidden Instructions

High
Category
Prompt Injection
Confidence
60% confidence
Finding

Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.

Content

Scanner excerpt · references/sample-roadmap.md (reported line 223)May include surrounding context.

md
### 溧阳中科海钠科技

溧阳中科海钠这件做的是层状氧化物的表面包覆,但包覆相的选择不是单一氧化物。化学式写作NaₓCuᵧFe₂MnₐM_bO₂⁠₊δ/(Z_cO_s + Y_rXO_t),分母部分表示两种包覆材料共存:Z_cO_s与Y_rXO_t。这是复合包覆而非单层包覆。层状氧化物正极的痛点在高电压或深度脱钠状态下的表面副反应与结构塌陷,包覆层需要同时满足隔绝电解液与允许钠离子通过两个条件。单一包覆物往往顾此失彼。该专利把两种包覆材料并列写入通式,表明方案要求的是协同包覆效果,而非某一特定化合物的作用。制备上采用一步或多步混合后高温烧结,工艺简洁,易于均匀包覆。与北京理工大学那件从基体成分改性切入不同,这件从表面工程切入,且以复合包覆为限定方式。层状氧化物路线在铜铁锰体系上的成本优势已被关注,表面处理是这一体系走向实用化的必经环节。该专利的辨识度在于用通式结构把两种包覆材料并列限定,形成对具体包覆物选择的宽覆盖[32]。企业研发管理者可将其视为层状氧化物表面工程的一个参考坐标:包覆设计正在从单一材料向复合体系演进。

### 北京理工大学

Hidden Instructions

High
Category
Prompt Injection
Confidence
60% confidence
Finding

Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.

Content

Scanner excerpt · references/sample-roadmap.md (reported line 231)May include surrounding context.

md
### 江苏正力新能电池技术:碳包覆与导电复合

江苏正力新能这件把MXene与O3相层状氧化物做成异质结。O3相成分为NaNiₐFe_bMn_cO₂,a+b+c=1;MXene通式为Mₙ⁠₊⁠₁XₙTₓ。两种材料通过固相烧结复合,在异质界面处形成强耦合。MXene的二维层状结构与高电子电导率,使其在复合后可能同时承担导电骨架与界面稳定层的角色。异质结的提法值得注意:它意味着两种材料在界面处形成了明确的能带或晶格匹配关系,而非简单的物理混合。O3相层状氧化物在循环中面临阳离子混排与相变问题,界面处的强耦合可能抑制表面相变的发生。这件专利与溧阳中科海钠那件都做层状氧化物的表面/界面处理,但路径不同:一个用无机氧化物包覆,一个用二维导电材料成结。两者并不互斥——包覆与异质结可以叠加,但彼此独立的专利布局表明,层状氧化物的界面工程尚未收敛到单一方案[34]。MXene在钠电正极中的应用整体处于较早期阶段,该专利的价值在于提供了一个可检验的复合结构,而非给出确定性的性能结论。

## 10、外部视角:产业化驱动与关键瓶颈

Code capability 'env' detected in scripts/patentmax_roadmap.py but not covered by declared permissions.

High
Category
MCP Least Privilege
Confidence
75% confidence
Finding

The skill uses 'env' capability that is not covered by the declared 'allowed-tools' preapproval. Review the host runtime's actual enforcement policy; runtime permissions are unknown unless separately verified, and this declaration alone does not establish a permission ceiling.

Content

No source excerpt is available for this finding.

Code capability 'network' detected in scripts/patentmax_roadmap.py but not covered by declared permissions.

High
Category
MCP Least Privilege
Confidence
75% confidence
Finding

The skill uses 'network' capability that is not covered by the declared 'allowed-tools' preapproval. Review the host runtime's actual enforcement policy; runtime permissions are unknown unless separately verified, and this declaration alone does not establish a permission ceiling.

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 62)May include surrounding context.

不需要密钥,装上就能用。 第一次运行时脚本自动领取一份免费试用:¥15 额度、30 天有效,存在本机 ~/.patentmax/。

试用用完后,在 api.ip930.com 注册(再送 ¥50 体验额度)并创建密钥,设成环境变量即可,设了就不再用试用:

bash
export PATENTMAX_API_KEY="pm_live_你的密钥"

The workflow provides no privacy warning before submitting potentially confidential patent-planning inputs to PatentMax.

Medium
Category
Output Handling
Confidence
95% confidence
Finding

The documented workflow encourages providing a context file containing potentially sensitive R&D plans, unpublished inventions, budgets, and confidentiality boundaries to an external patent-analysis service without an explicit privacy warning or informed-consent step. The patent-planning context increases the sensitivity of these inputs, and the examples use --yes; however, the README alone does not establish exactly which fields the script transmits or how the provider handles them.

Content

No source excerpt is available for this finding.

Error-handling instructions mandate verbatim Chinese responses without a language choice.

Medium
Category
Output Handling
Confidence
97% confidence
Finding

The skill explicitly identifies error messages as Chinese and requires the agent to relay them unchanged. This forces Chinese user-facing output without user opt-in or a documented justification for a language restriction.

Content

No source excerpt is available for this finding.

The materials-file workflow lacks a warning that supplied materials are sent to the external PatentMax service.

Medium
Category
Output Handling
Confidence
93% confidence
Finding

The materials-file instructions encourage supplying a local document without explicitly stating that its contents will be submitted to the external PatentMax service. In a patent-planning workflow, those materials may contain unpublished inventions, trade secrets, or confidential business plans, so this omission creates a meaningful risk of unintended third-party disclosure; the provided file does not establish the script's actual upload behavior or the service's data-handling protections.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% 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 · references/faq.md (reported line 125)May include surrounding context.

实在没有就直接发 HTTP 请求:

bash
curl -s -X POST "https://api.ip930.com/api/v1/reports" \
  -H "Authorization: Bearer $PATENTMAX_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: 自定义唯一串" \

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 · references/faq.md (reported line 125)May include surrounding context.

实在没有就直接发 HTTP 请求:

bash
curl -s -X POST "https://api.ip930.com/api/v1/reports" \
  -H "Authorization: Bearer $PATENTMAX_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: 自定义唯一串" \

Ae4

Medium
Category
analysis-evasion
Confidence
80% confidence
Finding

Suspicious Unicode normalization or mixed-script content

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% 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 · references/faq.md (reported line 25)May include surrounding context.

md
Windows PowerShell 用 `$env:PATENTMAX_API_KEY="..."`,CMD 用 `set PATENTMAX_API_KEY=...`。`export` 只对当前终端有效,换窗口要重设。

密钥在 https://api.ip930.com/features/api-platform 自助创建。

## 400 参数不正确

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% 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 · scripts/patentmax_roadmap.py (reported line 116)May include surrounding context.

python
# 密钥版(TRIAL_SOURCE 为空):必须设 PATENTMAX_API_KEY,不领试用。
TRIAL_SOURCE = "patent-portfolio-planning"
TRIAL_GROUP = "roadmap"
TRIAL_REGISTER_URL = "https://api.ip930.com/features/api-platform"
KEY_FROM_TRIAL = False

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% 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 · scripts/patentmax_roadmap.py (reported line 195)May include surrounding context.

python
if API_KEY:
        return API_KEY
    if not TRIAL_SOURCE:
        die("还没有设置 PatentMax 密钥。到 https://api.ip930.com/features/api-platform 注册(新用户送 ¥50 体验额度)"
            "并创建密钥,设成环境变量 PATENTMAX_API_KEY=pm_live_xxx 后,再运行同一条命令。", EXIT_INPUT)
    API_KEY = _load_trial_key() or _claim_trial_key()
    KEY_FROM_TRIAL = True

Tainted flow: 'path' from os.environ.get (line 171, credential/environment) → open (file write)

Medium
Category
Data Flow
Confidence
65% confidence
Finding

Data from a source is assigned to a variable that is later passed to a sink, creating a variable-mediated taint flow.

Content

Scanner excerpt · scripts/patentmax_roadmap.py (reported line 174)May include surrounding context.

python
path = _trial_key_path()
    try:
        os.makedirs(os.path.dirname(path), exist_ok=True)
        with open(path, "w", encoding="utf-8") as fh:
            json.dump({
                "api_key": key,
                "expires_at": data.get("expires_at"),

Tainted flow: 'operation' from os.environ.get (line 582, credential/environment) → open (file write)

Medium
Category
Data Flow
Confidence
65% confidence
Finding

Data from a source is assigned to a variable that is later passed to a sink, creating a variable-mediated taint flow.

Content

Scanner excerpt · scripts/patentmax_roadmap.py (reported line 360)May include surrounding context.

python
record.update({k: v for k, v in fields.items() if v is not None})
    try:
        os.makedirs(os.path.dirname(_task_path(operation)), exist_ok=True)
        with open(_task_path(operation), "w", encoding="utf-8") as fh:
            json.dump(record, fh, ensure_ascii=False, indent=2)
    except OSError:
        pass

Tainted flow: 'path' from os.environ.get (line 556, credential/environment) → open (file write)

Medium
Category
Data Flow
Confidence
65% confidence
Finding

Data from a source is assigned to a variable that is later passed to a sink, creating a variable-mediated taint flow.

Content

Scanner excerpt · scripts/patentmax_roadmap.py (reported line 557)May include surrounding context.

python
docx = try_download(f"/api/v1/reports/{urllib.parse.quote(report_id)}/files/report.docx")
    if docx:
        path = os.path.abspath(os.path.join(target, f"{REPORT_NAME}_{stem}.docx"))
        with open(path, "wb") as handle:
            handle.write(docx)
        files[REPORT_NAME] = path
    else:

Tainted flow: 'full' from os.environ.get (line 562, credential/environment) → open (file write)

Medium
Category
Data Flow
Confidence
65% confidence
Finding

Data from a source is assigned to a variable that is later passed to a sink, creating a variable-mediated taint flow.

Content

Scanner excerpt · scripts/patentmax_roadmap.py (reported line 563)May include surrounding context.

python
else:
        out["docx_note"] = "这次没有生成 Word 报告,结论以下面的摘要和完整结果为准。"
    full = os.path.abspath(os.path.join(target, f"{REPORT_NAME}_{stem}_结果.json"))
    with open(full, "w", encoding="utf-8") as handle:
        json.dump(payload, handle, ensure_ascii=False, indent=2)
    files["完整结果"] = full
    out["files"] = files

The documented 30-second command limit is contradicted by the implemented waiting behavior.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

L006–L008 explicitly promise that every command returns within 30 seconds to avoid blocking the host, but --wait invokes wait_for with a default timeout of 5,400 seconds (L640–L641). Even ordinary waits lack this guarantee: PATENTMAX_WAIT_SECONDS has no upper bound (L073), and each polling request can block for 60 seconds (L066, L241) without respecting the remaining deadline.

Content

No source excerpt is available for this finding.

Error-handling instructions require the agent to relay Chinese messages verbatim without offering a language choice.

Low
Category
Output Handling
Confidence
94% confidence
Finding

The die() docstring instructs the AI to repeat the Chinese user-facing explanation verbatim and not rephrase it; line 14 reinforces this requirement. This mandates Chinese error responses regardless of the user's language, with no explicit opt-in or documented locale-specific justification.

Content

No source excerpt is available for this finding.