ClawHub
Back to skill

Security audit

MOSI Transcribe Diarize 多说话人转写

Security checks for vulnerabilities and agentic risk

Overview

This skill is a straightforward audio transcription helper that sends user-provided audio to the declared MOSI Studio endpoint and writes transcript outputs locally.

Install only if you are comfortable sending the selected audio or video content to MOSI Studio for transcription. Avoid using it on confidential recordings unless that service is approved for your data, and choose output paths carefully because transcripts may contain sensitive speech content.

A.I.G

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

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Taint TrackingDirect Taint Flow, Variable-Mediated Taint Flow, Credential Exfiltration Chain
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (7)

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
93% confidence
Finding

The skill exposes meaningful capabilities—environment variable access, local file read/write, and outbound network use—without declaring an explicit permission or allowed-tools scope. That creates an authorization and transparency gap: a caller may not realize the skill can read local media and send content to a remote transcription service, increasing the chance of unintended data disclosure or misuse.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill instructs users to submit audio by URL, local file, or base64/data URL to a remote API endpoint, but it does not clearly warn that the audio content—and potentially sensitive local recording data—will be transmitted off-box to a third-party service. In a transcription/meeting-recording context, this can expose private conversations, personal data, or confidential business information without informed consent.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

This code accepts local audio via --file, converts it to a data URL, and sends the full audio payload to https://studio.mosi.cn/v1/audio/transcriptions. Although the network destination is constrained, the script does not present a confirmation prompt, user-facing disclosure, or inline warning that local audio data will be uploaded to a remote service.

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 · scripts/transcribe.py (reported line 101)May include surrounding context.

python
}

    try:
        r = requests.post(args.endpoint, headers=headers, json=payload, timeout=args.timeout)
    except requests.RequestException as e:
        fail(f"request failed: {e}")

Tainted flow: 'data' from requests.post (line 106, network input) → pathlib.Path.write_text (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/transcribe.py (reported line 116)May include surrounding context.

python
out_path = Path(args.out)
    if out_path.parent and not out_path.parent.exists():
        out_path.parent.mkdir(parents=True, exist_ok=True)
    out_path.write_text(json.dumps(data, ensure_ascii=False, indent=2), encoding="utf-8")

    segs = normalize_segments(data)

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
89% confidence
Finding

The file presents the skill description and operating instructions entirely in Chinese, including user-facing guidance, without indicating that language is optional or region-specific by design. Under the stated policy, forcing a specific language without user opt-in is a natural-language policy concern.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
79% confidence
Finding

The script writes the API response, per-segment transcript text, and speaker-grouped transcript text to disk, potentially creating multiple files containing sensitive speech content. There is no warning, confirmation, or explanatory comment/docstring disclosing that transcription results will be persisted locally.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.