Back to skill

Security audit

Agency Workflow Starter

Security checks for vulnerabilities and agentic risk

Overview

This is a legitimate n8n lead-routing template, but it needs review because it exposes a public webhook that can write unvalidated personal lead data into Google Sheets and Slack while overstating production readiness.

Install only after reviewing and hardening the n8n workflow. Add webhook authentication or upstream protection, validate and length-limit all fields, neutralize spreadsheet formulas, reduce what is posted to Slack, confirm privacy/consent and retention requirements, use least-privilege credentials, and replace the placeholder Dedup Check with a real CRM/sheet lookup or upsert before using it for a client.

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (2)

T09 · Insecure Skill Coding Practices

Warning
Location
lead-router.json:20
Finding

Untrusted Webhook Data Is Forwarded to Google Sheets and Slack Without Validation

Content
View full analysis
Remediation
View remediation
`, and `&` where required. - Prevent untrusted content from generating user, channel, or broadcast mentions. - Disable automatic parsing or link unfurling if supported by the node. 5. Authenticate webhook requests using an HMAC signature, shared secret, or trusted upstream authentication mechanism. 6. Add rate limiting, request-size limits, and replay protection at the reverse proxy, API gateway, or upstream form service. 7. Avoid retaining the complete raw request unless it is operationally necessary. If retained, apply the same validation and data-retention controls to `raw`. ]]>

T09 · Insecure Skill Coding Practices

Note
Location
lead-router.json:48
Finding

Duplicate Detection Is a Pass-Through Check Rather Than an Actual Lookup

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (8)

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 23)May include surrounding context.

md
- `lead-router.json` — n8n workflow export, import directly

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 32)May include surrounding context.

md
- `lead-router.json` — n8n workflow export, import directly

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · setup-notes.md (reported line 22)May include surrounding context.

md
**Airtable:**
- Credential type: `Airtable API`
- Needs: Airtable API key (Settings → Developer Hub → Personal access tokens)
- Base ID and Table name set in the Airtable node parameters

**HubSpot:**

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The README explicitly describes sending inbound lead data to multiple third-party services such as CRMs, Slack, Google Sheets, and Airtable, but provides no warning about handling personal data, consent, retention, or regulatory obligations. In a lead-capture workflow, this omission can cause operators to deploy processing of personal information without appropriate safeguards, increasing the risk of privacy violations and accidental data exposure.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The description promotes connecting the workflow to a client's system and handling inbound leads without warning about credential exposure, privacy obligations, tenant separation, or the sensitivity of lead data. In this context, omission of user-facing cautions is risky because the skill is explicitly marketed for fast client delivery, which may encourage rushed deployment into production systems with insufficient review.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The usage trigger phrases are broad enough to activate on common automation requests such as lead routing or inbound handling, which can cause the skill to be invoked outside a narrowly intended context. Because the skill then directs the user toward importing a workflow that connects to client systems, overbroad matching increases the chance of unintended deployment guidance being surfaced without adequate scoping, review, or safety checks.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The workflow forwards user-supplied lead data including name, email, company, phone, message, source, score, and bucket to Google Sheets and Slack, but nothing in the skill indicates consent capture, notice, minimization, or conditional routing based on privacy preferences. This creates a real privacy and compliance risk because personally identifiable information is disclosed to third-party services and internal channels automatically from a public webhook.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

This markdown file documents sending workflow data to Google Sheets, Airtable, HubSpot, and optionally Slack, but it does not warn that lead data will be transmitted to third-party systems or that Slack alerts may expose lead information in channels. For markdown files, SQP-2 applies when descriptions omit warnings about behaviors that could affect user data, privacy, or system integrity.

Content

No source excerpt is available for this finding.