ClawHub
Back to skill

Security audit

viewprinter-social-manager

Security checks for vulnerabilities and agentic risk

Overview

This skill is a clearly scoped ViewPrinter social-media manager with sensitive but disclosed posting, scheduling, media, account, and group-management capabilities.

Install only if you want an agent to operate your connected ViewPrinter account. Review requests carefully before allowing public posting, media deletion, or group membership changes, since those actions can affect real accounts and some cannot be recalled.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (6)

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 · evals/evals.json (reported line 137)May include surrounding context.

json
"scheduling"
      ],
      "prompt": "The video, caption, @example account and Friday September 18, 2026 at 9am America/Chicago are approved. For this post, enable the own-brand promotional label and AI-generated label; no paid-partnership label. Schedule it.",
      "expected_output": "uses the existing approval without asking for it again",
      "assertions": [
        "uses the existing approval without asking for it again",
        "resolves the absolute time and account id",

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 · evals/evals.json (reported line 139)May include surrounding context.

json
"scheduling"
      ],
      "prompt": "The video, caption, @example account and Friday September 18, 2026 at 9am America/Chicago are approved. For this post, enable the own-brand promotional label and AI-generated label; no paid-partnership label. Schedule it.",
      "expected_output": "uses the existing approval without asking for it again",
      "assertions": [
        "uses the existing approval without asking for it again",
        "resolves the absolute time and account id",

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 · evals/evals.json (reported line 248)May include surrounding context.

json
"scheduling"
      ],
      "prompt": "The video, caption, @example account and Friday September 18, 2026 at 9am America/Chicago are approved. For this post, enable the own-brand promotional label and AI-generated label; no paid-partnership label. Schedule it.",
      "expected_output": "uses the existing approval without asking for it again",
      "assertions": [
        "uses the existing approval without asking for it again",
        "resolves the absolute time and account id",

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 · evals/evals.json (reported line 250)May include surrounding context.

json
"scheduling"
      ],
      "prompt": "The video, caption, @example account and Friday September 18, 2026 at 9am America/Chicago are approved. For this post, enable the own-brand promotional label and AI-generated label; no paid-partnership label. Schedule it.",
      "expected_output": "uses the existing approval without asking for it again",
      "assertions": [
        "uses the existing approval without asking for it again",
        "resolves the absolute time and account id",

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The rule documents that groups_save replaces the full membership and that getting it wrong can silently empty a group, but it does not instruct the agent to warn the user or seek confirmation before making such a destructive change. In this skill, group membership controls where public posts may be sent, so an unintended overwrite can cause operational disruption, missed publishing targets, or accidental exclusion of accounts without any immediate error.

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 · references/rules/scheduling.md (reported line 45)May include surrounding context.

md
label, omit its parameter rather than inheriting `true` from an example,
template, prior post, or draft. Check saved options before releasing a draft;
if the tools cannot inspect or change them, resolve that before sending it.
Do not add disclosure wording to captions or overlays without approval either.

If current platform rules require a disclosure, explain the requirement and
ask how the user wants to proceed. Keep that destination held until resolved;