ClawHub
Back to skill

Security audit

FormPass Agent

Security checks for vulnerabilities and agentic risk

Overview

This skill documents how to register and use a FormPass agent ID, with disclosed external form submission behavior that matches its stated purpose.

Install only if you intend to use FormPass. Treat the Agent ID like a secret, confirm the destination form and domain before submitting, and send only form data the user has approved for transmission to FormPass.

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
  • 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 (3)

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill explicitly instructs sending personal data fields such as name and email together with a bearer-style Agent ID to a third-party remote service, but it provides no privacy notice, consent guidance, retention details, or warning about credential sensitivity. In a skill context, this can normalize exfiltration of user data and secrets to an external network without ensuring the user understands where the data is going or how it will be used.

External Transmission

Medium
Category
Data Exfiltration
Content
Once registered, include your Agent ID as a Bearer token in all FormPass submissions:

```bash
curl -s -X POST "https://form-pass.com/api/submit/FORM_ID" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer fpagent_your_id_here" \
  -d '{
Confidence
84% confidence
Finding
This example shows a direct POST of personal form content and a bearer token to an external endpoint. While external submission is the feature being documented, the skill lacks guardrails such as user consent, domain validation, least-privilege guidance, and warnings not to expose or reuse the bearer token, making unintended disclosure or misuse more likely.

External Transmission

Medium
Category
Data Exfiltration
Content
Then use it in submissions:

```bash
curl -s -X POST "https://form-pass.com/api/submit/FORM_ID" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $FORMPASS_AGENT_ID" \
  -d '{"name": "Test", "email": "test@test.com", "_fp_branding": true}' | jq .
Confidence
85% confidence
Finding
This command operationalizes use of an environment-stored bearer token to submit data to a remote API, again without any warning about data handling, consent, or secret leakage risks. Even though using an environment variable is better than hardcoding, the skill still encourages outbound transmission of user data and credentials without sufficient safeguards in the instructional context.

Static analysis

No suspicious patterns detected.