Back to skill

Security audit

Pet Separation Anxiety Relief (Owner Away) | 宠物分离焦虑舒缓(主人离家时)

Security checks for vulnerabilities and agentic risk

Overview

ClawScan marked this skill suspicious because it reports credential-handling and privacy risks, and its visible functions do not clearly match the advertised pet-monitoring and comfort actions.

Treat this as a review-level warning and consider not installing until you can verify that the account and credential handling is necessary and trusted, understand where video and credentials are sent, and confirm that the advertised pet monitoring and comfort actions are actually provided. The scan found meaningful risks but does not establish malicious intent.

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/smyx_separation_anxiety_relief_analysis.py
Finding

CLI identity initialization discards the upstream sender ID and reuses a shared account

Content
View full analysis

Impact

Reports generated for one sender are associated with the same cloud account used by other senders. A different sender's history-list request therefore uses the same credentials and can disclose the shared account's report records and report links. This does not establish access to arbitrary unrelated cloud accounts.

Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
Findings (26)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

The supplied code's implemented purpose is local database and user-account management, materially different from the declared pet-anxiety functionality. Database access could support a larger application, but this chunk contains no pet-specific data models or integration connecting its user-management capabilities to the stated purpose. This assessment is limited to the supplied chunk; it does not establish that the broader skill lacks the advertised functionality. No network transmission or token exfiltration is shown.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding

The supplied code implements a generic video-analysis CLI, not the declared end-to-end pet-anxiety detection and relief system. It forwards a file path or URL to skill.get_output_analysis and offers a separate history-list operation. Pet-specific analysis could exist in the imported implementation, which was not supplied, but this chunk contains no real-time monitoring loop, anxiety-threshold logic, or comfort-device orchestration. The assessment is limited to the visible code and does not assume malicious behavior or undisclosed behavior inside dependencies.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding

The supplied code does not implement the declared pet-anxiety detection and relief workflow. It also does not actually execute the commented-out external command. Imported base classes and services are not provided, so their capabilities cannot be assessed; this mismatch finding is limited to the visible code.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
94% confidence
Finding

The supplied code implements a generic video-analysis reporting client, materially narrower and different from the declared autonomous pet-monitoring and relief system. Video submission and report formatting could support the declared purpose, but the core real-time monitoring and device-control workflow is absent. The analysis API implementation is not supplied, so pet-specific processing inside that service cannot be ruled out; however, this chunk provides no evidence of it or of the promised automatic comfort actions.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
92% confidence
Finding

Generic file, JSON, polling, and HTTP helpers could legitimately support the declared skill and are not mismatches by themselves. However, this chunk also implements an undisclosed account and billing integration: it resolves or creates user identities, submits them to an external login/registration endpoint, persists returned credentials, and adds authentication headers to requests without restricting absolute URLs to the configured service. These behaviors go beyond the described pet-anxiety workflow. No camera analysis, anxiety detection, or comfort-device control appears in this chunk, although their absence from a shared utility file alone would not establish a mismatch. The finding is limited to the supplied code and does not establish malicious intent or actual credential disclosure.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
87% confidence
Finding

The code is related to the declared pet-anxiety use case, but its demonstrated functionality is video analysis and report retrieval, not the autonomous monitoring-and-relief workflow described. Identity resolution, history retrieval, and file output are supporting implementation details, not unrelated capabilities. The imported skill implementation is not supplied, so additional monitoring or device-control behavior inside that dependency cannot be verified.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
83% confidence
Finding

Generic AI-analysis configuration could support the declared purpose, and missing monitoring logic in a configuration-only chunk is not itself evidence of a mismatch. However, DETAIL_EXPORT_URL explicitly targets '/health/order/api/getReportDetailExport?id=', an inconsistent health-report resource with no stated connection to pet separation-anxiety monitoring. This chunk only configures that endpoint; it does not demonstrate an actual request, report download, or data disclosure.

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: "smyx-separation-anxiety-relief-analysis"
description: "AI-powered pet separation anxiety detection & relief when the owner leaves home. Real-time monitoring via smart camera detects typical anxiety signs—continuous vocalization, pacing, scratching doors/windows, destructive chewing. When anxiety reaches preset thresholds, the system auto-triggers comfort actions (play owner's pre-recorded voice, dispense treats via smart feeder, activate interactive toys) to reduce anxiety and destructive behavior, improving pet welfare. Scenarios: pet households (especially office workers / frequent travelers), pet boarding centers. | 通过智能

The HTTP helper can send platform credentials to arbitrary caller-supplied destinations.

High
Category
Excessive Agency
Confidence
99% confidence
Finding

The helper accepts arbitrary absolute HTTP or HTTPS URLs and automatically attaches platform access tokens, an API key, and an authorization token without checking the destination. An attacker who influences a request URL can receive those credentials, while an HTTP destination also exposes them in transit; compromised credentials could expose account data or connected-device capabilities.

Content

No source excerpt is available for this finding.

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 video-analysis workflow lacks a privacy warning about sending household footage to an external API.

Medium
Category
Output Handling
Confidence
90% confidence
Finding

Although API use is mentioned, the workflow does not clearly warn that local household video and audio are transmitted to an external service or obtain informed authorization for that disclosure. Pet-camera recordings can expose residents, private conversations, home interiors, and occupancy patterns, making the omission privacy-sensitive.

Content

No source excerpt is available for this finding.

The --export-env-only option promises not to run analysis, but the program still invokes video analysis.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

The parser accepts --export-env-only and explicitly promises not to analyze, but main never checks the flag before identity initialization or analysis. With an input supplied, a user relying on this non-executing mode can unintentionally trigger external processing of private footage and potentially incur charges.

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 that the file will be uploaded.

Medium
Category
Output Handling
Confidence
92% confidence
Finding

The implementation reads the complete local video and passes its bytes to the analysis API without an upload notice or an authorization check in this path. For indoor pet-camera footage, unexpected external disclosure can expose household activity and recorded conversations even though video analysis itself is legitimate.

Content

No source excerpt is available for this finding.

The --export-env-only option promises not to run analysis, but the program ignores it and executes analysis normally.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

The --export-env-only flag is parsed but never consulted, so providing it with a valid input still invokes analysis despite the explicit non-execution promise. This can cause unintended external processing of sensitive video, and identity initialization also occurs before any mode-specific safeguard.

Content

No source excerpt is available for this finding.

ai_chat claims to invoke the analysis agent, but the invocation is disabled and no analysis result can be returned.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

The agent invocation is commented out, and accessing stderr on an empty dictionary always raises an exception that is swallowed, causing ai_chat to return None. If monitoring relies on this helper, analysis and subsequent comfort actions can silently stop; the supplied source does not establish whether the pet-monitoring workflow actually calls it.

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
96% confidence
Finding

The generically named open helper always uses mode 'w', which immediately erases an existing file's contents. There is no confirmation, operation log, docstring, or comment explaining this destructive behavior; the exception handler does not disclose successful truncation.

Content

No source excerpt is available for this finding.

HTTP requests can silently register a remote account and persist authentication tokens locally.

Medium
Category
Output Handling
Confidence
90% confidence
Finding

A generic HTTP request can trigger a separate remote login or registration with silent=1 and register=1, then persist returned authentication tokens locally. This couples ordinary requests to identity disclosure and persistent account creation without an explicit opt-in in this path, an unexpected privacy side effect for a pet-monitoring skill.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
80% 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 · 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"):

Loading configuration silently creates directories and a YAML file when the file is missing.

Low
Category
Output Handling
Confidence
90% confidence
Finding

YamlUtil.load performs filesystem writes despite being a loading operation, without a warning, log, confirmation, or explanatory comment. BaseEnum.init_subclass invokes this loader during class definition (L052–L058), so importing this module can create configuration files without an explicit save request.

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: