ClawHub
Back to skill

Security audit

Session Log Analyzer

Security checks for vulnerabilities and agentic risk

Overview

The skill does what it says: it analyzes local session logs, creates a PDF report, and can manually upload part of that report to Notion.

Installers should review the generated PDF before running Notion sync, because the report can contain session IDs, errors, and other log-derived details that will be sent to Notion. Use a Notion integration token scoped only to the intended database.

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
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (7)

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

If the underlying skill does not actually perform the stated log analysis and PDF generation, while still relying on external Notion API access and environment-based credentials without declared permissions, users may be misled into granting secrets and data access under false assumptions. Description-behavior mismatch is dangerous because it undermines informed consent and can conceal data exfiltration or other unintended operations behind a benign-looking skill description.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
93% confidence
Finding

If the underlying skill does not actually perform the stated log analysis and PDF generation, while still relying on external Notion API access and environment-based credentials without declared permissions, users may be misled into granting secrets and data access under false assumptions. Description-behavior mismatch is dangerous because it undermines informed consent and can conceal data exfiltration or other unintended operations behind a benign-looking skill description.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
90% confidence
Finding

The skill declares access to environment variables and file inputs but does not define any explicit tool scope or permission boundaries. In a skill that handles logs and uses external-service credentials, this increases the chance of overbroad access and makes it harder for operators to understand or constrain what data the skill may read or expose.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill says it syncs report content to Notion but does not clearly warn that report text is transmitted to an external third-party service. Because session logs and derived reports may contain sensitive operational data, users could unintentionally disclose internal information, error details, or identifiers outside their local environment.

Content

No source excerpt is available for this finding.

Unsafe Defaults

Medium
Category
Tool Misuse
Confidence
75% confidence
Finding

Tool defaults are unsafe or overly permissive (e.g. disabled TLS verification, no authentication, world-writable permissions). Unsafe defaults widen the attack surface.

Content

Scanner excerpt · scripts/sync_to_notion.py (reported line 10)May include surrounding context.

python
NOTION_DATABASE_ID = os.environ.get("NOTION_REPORTS_DB_ID")

def sync_report_to_notion(pdf_path):
    notion = Client(auth=NOTION_API_KEY)
    
    # Extract text from PDF
    reader = PDFReader(pdf_path)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The script extracts text from a local PDF report and uploads it into Notion, which is an external SaaS platform, without any consent prompt, data-classification check, or redaction step. Because this skill analyzes session logs, the report may contain sensitive prompts, secrets, personal data, or internal operational details, making silent exfiltration to a third party a real privacy and security risk.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
89% confidence
Finding

The changelog text is written in Chinese while the rest of the skill documentation is in English, with no indication that the skill is region-specific or that language choice is intentional. This creates a locale inconsistency that can violate language policy when no user opt-in or justification is provided.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.