Back to skill

Security audit

技术交底书 · PatentMax

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed PatentMax client for generating patent disclosure Word documents, with the main risk being that sensitive invention materials are uploaded to PatentMax for processing.

Install only if you are comfortable sending the selected patent, R&D, code summaries, and invention details to PatentMax for processing. Redact secrets or unnecessary core parameters, use a trusted `PATENTMAX_API_KEY`, and do not set `PATENTMAX_BASE_URL` to an unknown host.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Rogue AgentSelf-Modification, Session Persistence
  • 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
Findings (29)

Tainted flow: 'req' from os.environ.get (line 221, 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_disclosure.py (reported line 145)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 221, credential/environment) → urllib.request.urlopen (network output)

Critical
Category
Data Flow
Confidence
90% confidence
Finding

The generic request() function builds outbound authenticated requests using an environment-controlled BASE_URL and includes the bearer API key in the Authorization header. If an attacker can influence PATENTMAX_BASE_URL in the agent runtime, they can redirect all report contents and the API credential to an arbitrary server, causing credential theft and exfiltration of sensitive patent/R&D materials.

Content

Scanner excerpt · scripts/patentmax_disclosure.py (reported line 230)May include surrounding context.

python
req.add_header(key, value)

    try:
        with urllib.request.urlopen(req, timeout=TIMEOUT) as response:
            payload = response.read()
            return payload if raw else json.loads(payload.decode("utf-8"))
    except urllib.error.HTTPError as exc:

Tp4

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding

The skill metadata says a PatentMax API key is required, but the implementation reportedly can auto-claim and cache a free trial key. That changes the trust and billing model materially: users or operators may believe no remote account provisioning or credential acquisition will occur, while the skill may silently obtain service access and transmit sensitive patent/R&D materials off-host. The undocumented .pptx parsing support is lower risk by itself, but it reinforces that the declared capabilities do not fully describe the actual data-handling behavior.

Content

No source excerpt is available for this finding.

Lp1

High
Category
MCP Least Privilege
Confidence
75% confidence
Finding

The skill uses 'env' capability that is not listed in its permissions. This may indicate deceptive intent or missing permission declarations.

Content

No source excerpt is available for this finding.

Lp1

High
Category
MCP Least Privilege
Confidence
75% confidence
Finding

The skill uses 'network' capability that is not listed in its permissions. This may indicate deceptive intent or missing permission declarations.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
84% confidence
Finding

The README explicitly directs users to use an external API endpoint, confirming that information leaves the local environment. In a patent-disclosure skill, this is especially sensitive because uploaded materials may include trade secrets, unpublished inventions, source code, and experimental data; absent clear disclosure and safeguards, users may expose valuable confidential information to a third party.

Content

Scanner excerpt · README.md (reported line 57)May include surrounding context.

怎么用

在 api.ip930.com 创建密钥,设成环境变量:

bash
export PATENTMAX_API_KEY="pm_live_你的密钥"

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The README instructs users to provide files or entire project directories to a service backed by an external PatentMax API, but it does not clearly warn that sensitive R&D materials, source code, and unpublished invention details may be transmitted off-system. In this skill’s context, that omission is meaningful because the data involved is often highly confidential and disclosure before filing can create business, contractual, or patent-strategy risks.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill processes highly sensitive, often pre-filing R&D and patent materials, yet the upfront description does not clearly warn that those materials will be transmitted to a remote service. This undermines informed consent: users may provide confidential inventions, trade secrets, diagrams, or internal documents without understanding they are leaving the local environment. In the patent context, undisclosed external transmission is especially sensitive because novelty, confidentiality, and internal IP controls matter.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The file instructs that when the API key is missing, the script will return a Chinese prompt and the agent should relay it verbatim. This imposes a specific language on user-facing communication without any opt-in or alternative language path, which is a locale/language policy concern.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The document states that error messages are written for users in Chinese and must be repeated verbatim. This prevents adaptation to the user's preferred language and can violate language-choice expectations unless the skill is clearly limited to Chinese-only users.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
90% confidence
Finding

This finding points to the same curl-based submission flow to the external PatentMax API. Because the payload includes technical disclosure content, the skill materially enables outbound transfer of potentially highly sensitive invention details to a remote service.

Content

Scanner excerpt · references/faq.md (reported line 59)May include surrounding context.

bash
# 提交交底书
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: 任取一个唯一串" \

External Transmission

Medium
Category
Data Exfiltration
Confidence
90% confidence
Finding

This finding points to the same curl-based submission flow to the external PatentMax API. Because the payload includes technical disclosure content, the skill materially enables outbound transfer of potentially highly sensitive invention details to a remote service.

Content

Scanner excerpt · references/faq.md (reported line 59)May include surrounding context.

bash
# 提交交底书
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: 任取一个唯一串" \

External Transmission

Medium
Category
Data Exfiltration
Confidence
89% confidence
Finding

The 'idea_upgrade' example sends user idea text to the same external API. Early-stage invention ideas are commercially sensitive, so this constitutes a real confidentiality risk in the context of a patent-disclosure skill.

Content

Scanner excerpt · references/faq.md (reported line 66)May include surrounding context.

md
-d '{"report_type":"technical_disclosure","input":{"supplement":"材料正文……","title_hint":"案件名称"}}'

# 出方向
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: 任取一个唯一串" \

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The entire guide directs users in Chinese and includes prescriptive Chinese output terms such as '用文字写出流程' and named document sections, but nowhere offers users a language/locale option or explains that the skill is intentionally limited to Chinese-language patent drafting. Under the policy, forcing a specific language without user opt-in is a natural-language policy concern.

Content

No source excerpt is available for this finding.

Session Persistence

Medium
Category
Rogue Agent
Confidence
60% confidence
Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · scripts/patentmax_disclosure.py (reported line 11)May include surrounding context.

python
**每条命令最多跑 30 秒就返回。** WorkBuddy 这类宿主前台跑命令时,要等命令结束才显示输出,
AI 中途也插不了话;一条命令阻塞十几分钟,用户就十几分钟什么都看不到,宿主还可能把命令掐掉。
所以 create / ideas 提交后只等 15 秒就先返回,之后用 wait 接着等,每次最多 30 秒:

    退出码 0   这一步有结果了(wait 完成时已自动下载好 Word)
    退出码 10  还在生成:把输出里的「[进度]」行告诉用户,再运行一次 wait

Session Persistence

Medium
Category
Rogue Agent
Confidence
60% confidence
Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · scripts/patentmax_disclosure.py (reported line 11)May include surrounding context.

python
**每条命令最多跑 30 秒就返回。** WorkBuddy 这类宿主前台跑命令时,要等命令结束才显示输出,
AI 中途也插不了话;一条命令阻塞十几分钟,用户就十几分钟什么都看不到,宿主还可能把命令掐掉。
所以 create / ideas 提交后只等 15 秒就先返回,之后用 wait 接着等,每次最多 30 秒:

    退出码 0   这一步有结果了(wait 完成时已自动下载好 Word)
    退出码 10  还在生成:把输出里的「[进度]」行告诉用户,再运行一次 wait

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The code explicitly instructs that error messages are Chinese and should be relayed verbatim, and many CLI/help strings throughout the file are Chinese-only. This imposes a fixed language on users without opt-in, which matches the locale-policy violation criteria for natural-language policy issues.

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 7)May include surrounding context.

md
# 密钥版(TRIAL_SOURCE 为空):必须设 PATENTMAX_API_KEY,不领试用。
TRIAL_SOURCE = ""
TRIAL_GROUP = "disclosure"
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 · references/faq.md (reported line 74)May include surrounding context.

md
# 密钥版(TRIAL_SOURCE 为空):必须设 PATENTMAX_API_KEY,不领试用。
TRIAL_SOURCE = ""
TRIAL_GROUP = "disclosure"
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 · references/faq.md (reported line 78)May include surrounding context.

md
# 密钥版(TRIAL_SOURCE 为空):必须设 PATENTMAX_API_KEY,不领试用。
TRIAL_SOURCE = ""
TRIAL_GROUP = "disclosure"
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 · references/faq.md (reported line 82)May include surrounding context.

md
# 密钥版(TRIAL_SOURCE 为空):必须设 PATENTMAX_API_KEY,不领试用。
TRIAL_SOURCE = ""
TRIAL_GROUP = "disclosure"
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 · references/faq.md (reported line 84)May include surrounding context.

md
# 密钥版(TRIAL_SOURCE 为空):必须设 PATENTMAX_API_KEY,不领试用。
TRIAL_SOURCE = ""
TRIAL_GROUP = "disclosure"
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_disclosure.py (reported line 103)May include surrounding context.

python
# 密钥版(TRIAL_SOURCE 为空):必须设 PATENTMAX_API_KEY,不领试用。
TRIAL_SOURCE = ""
TRIAL_GROUP = "disclosure"
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_disclosure.py (reported line 182)May include surrounding context.

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

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
79% confidence
Finding

The README consistently instructs and exemplifies usage only in Chinese, including the recommended invocation phrase and all output descriptions. Under the policy, forcing a specific language without user opt-in can be a natural-language policy issue unless the locale limitation is explicitly documented and justified.

Content

No source excerpt is available for this finding.