Back to skill

Security audit

Workplace Phone Usage Smart Monitoring Skill | 职场玩手机智能监测技能

Security checks for vulnerabilities and agentic risk

Overview

The scanners rate this skill suspicious with high confidence, citing unexpected account and credential handling, remote processing of workplace media without clear disclosure, and an export-only option that may still run analysis.

Consider not installing until you have verified where monitoring media and identity data are sent, obtained appropriate authorization for remote processing and account creation, and confirmed the export-only option does not run analysis. The skillspector scan is partial because a referenced path was missing, so its findings do not provide complete evidence about that reference.

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (1)

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/phone_usage_monitoring_analysis.py
Finding

Sender identity is discarded, allowing different senders to reuse one cloud account

Content
View full analysis

Impact

A second sender can retrieve reports created through another sender's fallback account, exposing workplace monitoring results. New analyses are also associated with the same account. This finding applies to shared-sender integrations, not an intentionally single-user workspace.

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
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • YARA SignaturesMalware Match, Webshell Match, Cryptominer Match
Findings (23)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
91% confidence
Finding

Generic file, JSON, polling, and HTTP helpers can reasonably support the declared detection service and are not themselves mismatches. However, this chunk additionally implements an undeclared account-provisioning and credential-management workflow: it resolves an identity from local configuration or storage, sends it to a remote login endpoint with silent registration enabled, and persists returned tokens. Its request wrapper also attaches credentials to absolute URLs supplied by callers without checking their destination. These capabilities extend beyond the advertised monitoring behavior. The code does not establish that these methods are invoked by the full skill or that credentials are actually sent to an untrusted destination; the absence of vision logic in a shared utility file alone would not justify the finding.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
91% confidence
Finding

The supplied chunk does not implement the declared computer-vision workflow. Its only substantive method is an incomplete agent-chat helper, and it does not actually execute the commented-out external command. Behavior inherited from the imported classes cannot be determined from this chunk, so this assessment is limited to the visible implementation.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
90% confidence
Finding

The supplied chunk's substantive functionality is user-account persistence, not computer vision or phone-use analytics. Generic database access could support the declared skill, but this implementation specifically manages account profiles and tokens, with no visible connection to detection results, usage events, durations, or frequencies. The mismatch is therefore the account-management purpose, not merely the presence of database plumbing. This assessment is limited to the supplied chunk; it does not show network transmission or establish what other skill files do.

Content

No source excerpt is available for this finding.

YARA rule 'agent_skill_mcp_tool_poisoning_metadata': MCP/tool metadata poisoning indicators in tool schemas or skill manifests [agent_skills]

High
Category
YARA Match
Confidence
80% confidence
Finding

YARA rule matched a hack tool or exploit indicator (offensive tools, reconnaissance, privilege escalation, or exploit frameworks).

Content

Scanner excerpt · SKILL.md (reported line 3)May include surrounding context.

md
---
name: "phone_usage_monitoring_analysis"
description: "Based on computer vision, automatically detects employees playing with phones during work hours, supports real-time video stream and image detection, counts the duration and frequency of phone usage, helps enterprises standardize office order, and improves work efficiency. | 职场玩手机智能监测技能,基于计算机视觉自动检测工作时间员工玩手机行为,支持视频流和图片实时检测,统计玩手机时长与频次,帮助企业规范办公秩序,提升工作效率"
version: "1.0.25"
license: "MIT-0"
---

# 📱 Workplace Phone Usage Smart Monit

Skill declares no tool scope ('permissions' or 'allowed-tools') but code capabilities were detected: env, file_read, file_write, network, shell.

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding

Without declared permissions the skill's intent is opaque and cannot be validated.

Content

No source excerpt is available for this finding.

The analysis workflow lacks a clear warning that employee monitoring media is processed by a remote API.

Medium
Category
Output Handling
Confidence
90% confidence
Finding

Although the documentation mentions APIs and cloud reports, it does not clearly disclose that selecting a local monitoring file causes its contents to leave the machine for remote processing. Employee footage can contain identifiable people and confidential workplace information, making this an important authorization and privacy boundary.

Content

No source excerpt is available for this finding.

The --export-env-only option promises not to run analysis, but execution proceeds with analysis anyway.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

The parser accepts --export-env-only and promises not to perform analysis, but the flag is never checked before identity initialization or analysis. When supplied with an input, this supposed non-analysis mode can initiate remote processing of sensitive employee footage contrary to the caller's explicit expectation.

Content

No source excerpt is available for this finding.

Whitespace Padding

Medium
Category
Prompt Injection
Confidence
70% confidence
Finding

Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.

Content

Scanner excerpt · skills/smyx_analysis/scripts/skill.py (reported line 28)May include surrounding context.

python
result_json = JsonUtil.parse(result_json_pure_text, result_json_pure_text)

        result_json_common_ai_response = result_json.get("commonAiResponse") if isinstance(result_json,
                                                                                           dict) else result_json
        if result_json_common_ai_response:
            result_json = result_json_common_ai_response

Whitespace Padding

Medium
Category
Prompt Injection
Confidence
70% confidence
Finding

Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.

Content

Scanner excerpt · skills/smyx_analysis/scripts/skill.py (reported line 33)May include surrounding context.

python
result_json = result_json_common_ai_response

        result_json_health_ai_response = result_json.get("healthAiResponse") if isinstance(result_json,
                                                                                           dict) else result_json
        if result_json_health_ai_response:
            result_json = result_json_health_ai_response

Local video contents are passed to the analysis API without disclosure of the upload.

Medium
Category
Output Handling
Confidence
93% confidence
Finding

The local-file branch reads the entire media file and passes its contents to the analysis API without a visible upload-disclosure or authorization check. This crosses a significant privacy boundary for workplace surveillance footage, even though remote analysis is a legitimate feature.

Content

No source excerpt is available for this finding.

The documented export-only mode still executes video analysis.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

The export-only option is parsed but never consulted, so providing an input still invokes video analysis. A caller relying on the advertised no-analysis behavior can inadvertently submit private footage for remote processing.

Content

No source excerpt is available for this finding.

save() is documented as creating a new record but can overwrite an existing record instead.

Medium
Category
Excessive Agency
Confidence
93% confidence
Finding

Any insertion exception triggers an update attempt by ID and then by username or realname, copying fields that can include credentials and identifiers. Consequently, a create operation involving conflicting input can overwrite an existing user's record instead of failing; exploitation depends on whether an untrusted caller can influence the supplied model.

Content

No source excerpt is available for this finding.

ai_chat documents invoking an analysis agent but never executes it or returns analysis.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

The docstring at L25–L34 claims the method invokes openclaw agent through subprocess and returns analysis results or a session ID. However, the subprocess invocation is commented out, and the empty dictionary assigned at L68 causes result.stderr to raise AttributeError; the catch block suppresses it and the method returns None. This contradicts the method's core documented behavior rather than merely changing an implementation detail.

Content

No source excerpt is available for this finding.

FileUtil.open silently truncates existing files without disclosing its destructive behavior.

Medium
Category
Output Handling
Confidence
97% confidence
Finding

The generically named open helper always uses write mode, which immediately erases an existing file's contents. No docstring, comment, confirmation, or user-facing log explains this behavior; exception tracing only reports failures.

Content

No source excerpt is available for this finding.

The request helper silently registers external service accounts beyond the declared phone-usage analysis functionality.

Medium
Category
Excessive Agency
Confidence
94% confidence
Finding

An ordinary request can automatically call /sys/phoneLogin with silent=1 and register=1 when credentials are missing, without a separate enrollment decision. This transmits identity information and can create a remote account, a consequential side effect not disclosed by the phone-monitoring description; legitimate backend integration does not remove the need for explicit provisioning consent.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
93% confidence
Finding

This POST sends the selected identity as both openId and mobile to the configured health-service login endpoint. The selection explicitly prefers ApiEnum.API_SECRET_KEY, so a configured API secret can be transmitted as ordinary registration data rather than kept in its intended authentication channel; the endpoint's trust and necessity are not established by the supplied source.

Content

Scanner excerpt · skills/smyx_common/scripts/util.py (reported line 561)May include surrounding context.

python
"source": ConstantEnum.DEFAULT__SKILL_HUB_NAME
            }
            try:
                _response = requests.post(_url, json=_data)
                if _response.status_code == 200:
                    _response_json = _response.json()
                    if _response_json and _response_json.get("success"):

HTTP requests can silently trigger remote account registration and local credential persistence.

Medium
Category
Output Handling
Confidence
93% confidence
Finding

The request helper can obtain remote credentials and persist token and openToken values through UserDao.save as an implicit consequence of sending a request. The credentials are placed in ordinary SQLite string columns without protection shown here, expanding the credential footprint and exposing them to anyone able to read the database.

Content

No source excerpt is available for this finding.

Configuration loading silently creates directories and writes missing configuration files.

Low
Category
Output Handling
Confidence
92% confidence
Finding

At L17–L20, YamlUtil.load creates parent directories and writes a YAML file when the requested file is absent, without a warning, log, docstring, or explanatory comment. BaseEnum.init_subclass calls this loader during class definition at L53 and L58, so merely importing the module can modify the filesystem without disclosure.

Content

No source excerpt is available for this finding.

Dynamic attribute access via getattr()

Low
Category
Dangerous Code Execution
Confidence
50% confidence
Finding

Dynamic getattr() with a non-literal attribute name can access arbitrary object attributes, potentially bypassing access controls.

Content

Scanner excerpt · skills/smyx_common/scripts/dao.py (reported line 243)May include surrounding context.

python
username = getattr(model, "username", None)
            if username:
                column_names = self.__model__.__table__.columns.keys()
                update_data = {key: getattr(model, key) for key in column_names if key != "username"}
                return self.update_by_username(username, **update_data)

            return None

Dynamic attribute access via getattr()

Low
Category
Dangerous Code Execution
Confidence
50% confidence
Finding

Dynamic getattr() with a non-literal attribute name can access arbitrary object attributes, potentially bypassing access controls.

Content

Scanner excerpt · skills/smyx_common/scripts/dao.py (reported line 353)May include surrounding context.

python
column_names = self.__model__.__table__.columns.keys()

            for key in column_names:
                value = getattr(model, key)
                setattr(instance, key, value)

            session.commit()

Dynamic attribute access via getattr()

Low
Category
Dangerous Code Execution
Confidence
50% confidence
Finding

Dynamic getattr() with a non-literal attribute name can access arbitrary object attributes, potentially bypassing access controls.

Content

Scanner excerpt · skills/smyx_common/scripts/dao.py (reported line 326)May include surrounding context.

python
if filters:
                for key, value in filters.items():
                    query = query.filter(getattr(self.__model__, key) == value)

            if offset:
                query = query.offset(offset)

Dynamic attribute access via getattr()

Low
Category
Dangerous Code Execution
Confidence
50% confidence
Finding

Dynamic getattr() with a non-literal attribute name can access arbitrary object attributes, potentially bypassing access controls.

Content

Scanner excerpt · skills/smyx_common/scripts/dao.py (reported line 441)May include surrounding context.

python
if filters:
                for key, value in filters.items():
                    query = query.filter(getattr(self.__model__, key) == value)

            return query.scalar()
        finally:

Dynamic attribute access via getattr()

Low
Category
Dangerous Code Execution
Confidence
50% confidence
Finding

Dynamic getattr() with a non-literal attribute name can access arbitrary object attributes, potentially bypassing access controls.

Content

Scanner excerpt · skills/smyx_common/scripts/util.py (reported line 30)May include surrounding context.

python
elif isinstance(cur, dict):
                        cur = cur.get(part, default)
                    else:
                        cur = getattr(cur, part)
                return cur
            except Exception:
                return default