ClawHub
Back to skill

Security audit

knowledge-graph

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed knowledge-graph API client that stores and queries user-authorized graph data on its platform with reasonable limits and secret-handling rules.

Install this only if you intend to use the AI Skills hosted knowledge-graph service and are comfortable storing selected entities, relations, and source URLs there. Do not place secrets or sensitive personal data in graph fields, and prefer the product-specific KNOWLEDGE_GRAPH_API_KEY if you want tighter key separation.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The skill activation description is very broad ('when users want to build a knowledge graph, query entity relations, or generate source-backed summaries') and does not define clear boundaries, exclusions, or preconditions. In an agentic system, this can cause over-invocation on loosely related prompts, unnecessarily exposing user data to the external platform and expanding the chance of unintended API calls or context leakage.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The description is written as a Chinese-only usage rule without stating whether other user languages are supported or how language preference should be handled. This can lead the agent to default to Chinese behavior even when the user did not request it, increasing the risk of user confusion, incorrect task execution, or accidental disclosure through mistranslation in security-relevant or source-sensitive workflows.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% 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 · references/HTTP-REQUESTS.md (reported line 10)May include surrounding context.

bash
IDEMPOTENCY_KEY="$(uuidgen | tr '[:upper:]' '[:lower:]')"

curl --fail-with-body --silent --show-error \
  -X POST "$API_ROOT/knowledge-graph/entity.upsert" \
  -H "Authorization: Bearer ${KNOWLEDGE_GRAPH_API_KEY:-$AI_SKILLS_API_KEY}" \
  -H "Content-Type: application/json" \

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

This markdown file contains user-facing operational instructions exclusively in Chinese, and there is no indication that users may select another language or that the skill is intentionally limited to a Chinese-speaking context. That can violate a language/locale policy requiring user opt-in or clear justification for locale restrictions.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.