ClawHub
Back to skill

Security audit

Test Skill 3

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed creative media automation skill that uses configured AI providers and local output files, with no hidden or destructive behavior found.

Install only if you are comfortable letting the skill send prompts, briefs, and selected media inputs to the configured AI providers and consume API credits. Avoid putting secrets or confidential campaign details in prompts, use dry-run for full pipelines first, and review output directories and task logs when handling sensitive work.

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
--argjson influence "$INFLUENCE" \
    '{text: $text, duration_seconds: $duration, loop: $loop, prompt_influence: $influence}')

  local endpoint="https://api.freepik.com/v1/ai/sound-effects"

  if [[ "$DRY_RUN" == "true" ]]; then
    json_output "$(json_build command="curl -s -X POST '${endpoint}' ..." provider="$SELECTED" model=sound-effects dry_run=true)"
Confidence
95% confidence
Finding
This script sends user-supplied prompts and generation parameters to an external Freepik API endpoint, which is a real external data transmission. In a headless agent skill, prompts may contain sensitive user or workspace data, so forwarding them to third-party providers creates privacy and data-handling risk if callers are not clearly informed or constrained.

External Transmission

Medium
Category
Data Exfiltration
Content
local api_key
  api_key=$(get_provider_key elevenlabs)

  local endpoint="https://api.elevenlabs.io/v1/sound-generation"

  local payload
  payload=$(jq -n \
Confidence
95% confidence
Finding
This endpoint definition is used to transmit prompts to ElevenLabs for sound generation, so it represents genuine outbound data flow to a third party. Given the skill's purpose as an automated creative-production pipeline, agents may pass through proprietary or sensitive text, making this more dangerous than a simple local utility if no guardrails exist.

Static analysis

No suspicious patterns detected.