Back to skill

Security audit

Reviews For AI Answers

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this skill benign because its checks found no reported security issues and the flagged item was considered purpose-aligned, though the overall evidence is incomplete.

The verdict is benign with medium confidence, not a guarantee of safety. Skillspector’s analysis was partial: it could not resolve several local path-like references, so their context was not fully assessed. Consider that limitation alongside the possible instruction-scope concern when deciding whether to install.

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
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

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 85)May include surrounding context.

md
the direct review link: the platform's review-collection link (G2, Capterra, Trustpilot invitations), the Google review link from the profile, the StoreKit review prompt on iOS, Google Play's In-App Review API, Amazon's "Request a Review" button.
   - Wording: neutral and short. "How has [product] worked for you? An honest review on [platform] helps other [buyers], good or bad." Never mention stars, suggest content or offer anything the platform bans.
   - Platform rules to apply, as checked in late 2026 (re-check each one before launch):
     - Google: no incentives, no gating, no staff review quotas, no asking for specific content, no pressure on premises.
     - Trustpilot: no incentives; invite every customer the same way at the same point in the journey, with non-leading wording.
     - Yelp: don't ask at all.
     - G2: incentives allowed within G2's cap, disclosed to G2 and never tied to sentiment; G2 labels incentivised reviews and bans segmenting out likely negative reviewers.