ClawHub
Back to skill

Security audit

f100k-puente

Security checks for vulnerabilities and agentic risk

Overview

This bridge skill is coherent and disclosed; it uses external services and private workspace context for the stated F100K workflow, with approval gates for spending and public actions.

Install only if you are comfortable with this skill reading the F100K workspace context and using connected third-party services for research, scraping, media generation, deployment, and Telegram delivery. Before approving paid actions or public posting, review previews and remember that Gemini image generation may send prompts and selected photos to Google.

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 (3)

Vague Triggers

Medium
Confidence
90% confidence
Finding
The description says to read this skill 'SIEMPRE antes de usar cualquier skill F100K' and whenever a skill mentions a wide range of tools, paths, or email delivery. This creates an expansive and ambiguous trigger scope without clear boundaries or negative examples, increasing the chance of unintended invocation.

External Transmission

Medium
Category
Data Exfiltration
Content
```bash
mkdir -p entregables/imagenes
curl -s -X POST \
  "https://generativelanguage.googleapis.com/v1beta/models/gemini-3.1-flash-image-preview:generateContent?key=$GEMINI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
Confidence
83% confidence
Finding
The skill instructs the agent to transmit user-provided content and prompts to an external third-party API using curl. In context, this can include personal photos from `cerebro/fotos/` and other workspace content, creating privacy and data-governance risk if sent without explicit, per-request consent and clear minimization rules.

Natural-Language Policy Violations

Low
Confidence
94% confidence
Finding
The instructions require 'prompt en inglés' as a fixed rule for image prompting. This is a language-policy constraint presented as mandatory, and the file does not offer the user a language choice or explain a documented regional/compliance need for enforcing English.

Static analysis

No suspicious patterns detected.