Back to skill

Security audit

Child Learning Behavior Analysis Tool | 孩子学习行为分析工具

Security checks for vulnerabilities and agentic risk

Overview

ClawScan marks this study video analysis skill suspicious because scanners report risks involving remote credentials, user identity, and behavior that may process videos despite an export-only choice; the review is partial because one reference is missing.

Consider the reported risks before installing, especially if videos or reports contain sensitive student information. Review the account registration and credential handling, confirm where videos are sent, and verify that export-only truly prevents analysis. Coverage is incomplete: a local reference could not be resolved, even though the listed components were scanned.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Findings (21)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
89% confidence
Finding

The instructions require silently reusing a local default identity when no upstream identity is supplied and automatically associating historical reports with that identity. On a workspace serving multiple people, this can mix or expose children's reports across users; however, the additional claims about credential-file transmission and payment-skill installation are not independently substantiated by this source.

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: "study-analysis"
description: "Conducts video analysis of learning behavior for children/students, identifies poor learning habits, provides structured analysis reports and family education improvement suggestions, focusing on learning habit cultivation and behavior correction. | A comprehensive tool designed to analyze video footage of children's and students' learning behaviors. It identifies poor study habits and provides structured analysis reports along with actionable suggestions for family education improvements. The tool is dedicated to fostering positive study habits and facilitating behavioral correc

Arbitrary absolute request URLs receive the skill's shared API credentials.

High
Category
Excessive Agency
Confidence
98% confidence
Finding

The helper accepts arbitrary absolute HTTP or HTTPS URLs and automatically attaches X-Access-Token, X-Api-Key, and Authorization without checking the destination origin. If an attacker can influence a caller's URL, credentials can be sent to an attacker-controlled server; HTTP destinations additionally expose them to network interception. Compromised credentials could expose sensitive student-analysis data, depending on their permissions.

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 documented export-only mode does not prevent video analysis.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

The parser advertises an export-only option that will not perform analysis, but the parsed flag is never checked. With a video input supplied, execution still initializes identity and calls the analysis workflow, potentially transmitting sensitive children's footage despite the user's explicit no-analysis selection.

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 submitted to the analysis API without disclosure of the upload.

Medium
Category
Output Handling
Confidence
90% confidence
Finding

The code reads the entire local video into a multipart file payload and passes it to the analysis API. No prompt, log, docstring, or comment explains that the user's video is being transmitted to a service; the existing comment only describes reading the file.

Content

No source excerpt is available for this finding.

The documented export-only mode does not prevent video analysis.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

The export-only option explicitly promises not to run analysis, yet its value is never consulted. Supplying an input alongside this flag still invokes identity resolution and video analysis, creating a risk of unintended remote processing of private footage.

Content

No source excerpt is available for this finding.

Configuration loading silently creates directories and files, including during module import.

Medium
Category
Output Handling
Confidence
94% confidence
Finding

YamlUtil.load creates parent directories and writes a YAML file when the requested path does not exist, without a prompt, log, docstring, or explanatory comment disclosing this side effect. BaseEnum.init_subclass invokes this loader at L053 and L058, so importing the module can modify the filesystem without an explicit save operation.

Content

No source excerpt is available for this finding.

The ai_chat docstring claims to invoke an analysis agent, but the implementation never invokes it.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

The docstring at L25–L35 states that ai_chat calls openclaw agent through subprocess and returns analysis results or a session ID. However, command construction and subprocess execution are commented out (L48–L66); the empty dictionary at L68 then causes an attribute-access exception that is caught at L81–L82, so the method returns None without performing the documented analysis.

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

Opening a path with mode 'w' immediately erases any existing contents, even before the caller writes anything. This generically named helper has no docstring, explanatory comment, confirmation, or action log disclosing that behavior; exception tracing does not disclose successful truncation.

Content

No source excerpt is available for this finding.

The request helper silently registers a remote account and persists its credentials, beyond the declared video-analysis workflow.

Medium
Category
Excessive Agency
Confidence
90% confidence
Finding

L550-L561 submit an identity to /sys/phoneLogin with both silent and register enabled. When tokens are unavailable, L589-L604 invoke this operation and save the returned user information and tokens locally; the manifest describes learning-behavior analysis, not automatic account enrollment and persistent identity management.

Content

No source excerpt is available for this finding.

HTTP requests can silently trigger remote registration and local storage of authentication tokens.

Medium
Category
Output Handling
Confidence
92% confidence
Finding

When authentication tokens are missing, the HTTP helper sends the current identity to a separate phoneLogin endpoint with registration enabled, then saves returned user information and tokens through UserDao. These account-creation and credential-persistence side effects are not disclosed by a comment, docstring, confirmation, or action log, and go beyond merely performing the caller's requested HTTP operation.

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"):

The example adds the generic phrase “历史报告” (“historical reports”) as an automatic trigger without requiring a child-learning context.

Low
Category
Trigger Abuse
Confidence
93% confidence
Finding

The generic historical-reports trigger is combined with instructions to automatically query the cloud and display results. An unrelated request could therefore retrieve sensitive child-learning reports without the user intending to access this service.

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