ClawHub
Back to skill

Security audit

HAGGIN

Security checks for vulnerabilities and agentic risk

Overview

This skill is a clearly disclosed HAGGIN API helper for seller-side price negotiations, with meaningful business access but explicit confirmation gates for actions that affect buyers or deals.

Install this only for agents you trust with HAGGIN seller data and a HAGGIN API key. Prefer a read-only key when reviewing offers, and use a full-access key only when you want the agent to create drafts or perform confirmed seller actions. Review proposed request bodies before approving buyer-facing actions or webhook changes.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (4)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The invocation description is unusually broad and includes generic phrasing like using the skill whenever users mention HAGGIN-related concepts or letting customers make an offer on pricing. This can cause the skill to activate in contexts where the user did not explicitly request negotiation or API access, increasing the chance of unnecessary access to sales data or unintended operational suggestions.

Content

No source excerpt is available for this finding.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
75% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 61)May include surrounding context.

md
**The seller decides. You prepare.**

Do freely, without asking:

- Read anything: Haggins, the inbox, negotiations, deals, deliveries.
- Interpret offers, rank them, explain them.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The markdown states that webhook endpoint creation and secret rotation responses include the webhook signing secret, but it does not warn users that this credential must be stored securely and treated as sensitive. Because this secret protects webhook authenticity, omitting a warning increases the risk of accidental leakage or unsafe handling.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
83% confidence
Finding

The documentation says webhook deliveries and webhook payloads can include payload, response_body, and buyer data including email when consented, but it does not warn that these fields may contain sensitive business or personal information. For markdown files, behaviors affecting privacy should be clearly disclosed so users understand the data exposure implications of enabling and inspecting webhooks.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.