ClawHub
Back to skill

Security audit

Live Search

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed live web-search helper that sends search terms through the host app’s local gateway and shows no hidden persistence, destructive behavior, or unrelated data access.

Install this if you are comfortable with the assistant using the host app’s search gateway for live web lookups. Keep queries concise, do not include secrets or confidential data, and verify important results at the linked sources.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

External Transmission

Medium
Category
Data Exfiltration
Content
PPID_VAL=$(python3 -c "import os; print(os.getppid())")
echo "[Assistant] Parent PID: $PPID_VAL"

curl -s -X POST http://localhost:$PORT/proxy/prosearch/search \
  -H 'Content-Type: application/json' \
  -d '{"keyword":"your search query"}'
```
Confidence
95% confidence
Finding
This skill instructs the agent to transmit user-derived search queries to a local authenticated gateway, which then proxies them to external web search infrastructure. Even though the HTTP target is localhost, the documented purpose is live public-web retrieval, so user prompts and derived keywords may leave the local trust boundary and expose sensitive user intent or data if the agent searches for confidential content.

External Transmission

Medium
Category
Data Exfiltration
Content
echo "[Assistant] Parent PID: $PPID_VAL"

# Basic
curl -s -X POST http://localhost:$PORT/proxy/prosearch/search \
  -H 'Content-Type: application/json' \
  -d '{"keyword":"latest AI news"}'
Confidence
94% confidence
Finding
The repeated curl examples reinforce the same behavior: sending user-derived content through a session-authenticated local proxy for external search. In context, this is the core function of the skill, but it still represents an external transmission risk because nothing in the skill prevents sensitive or internal information from being included in outbound queries.

Static analysis

No suspicious patterns detected.