ClawHub
Back to skill

Security audit

autoresearch-pro

Security checks for vulnerabilities and agentic risk

Overview

This skill gives bounded instructions for improving an existing artifact and includes clear limits around scope, reversibility, permissions, and external actions.

Install this if you want an agent to perform small, evidence-based revisions to existing artifacts. Review proposed edits and comparisons, especially when the target is important, but the package does not include executable code, credential use, hidden persistence, or automatic external actions.

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
  • 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
80% 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 37)May include surrounding context.

md
- Save the exact current version outside the active skill discovery path, or record
  a recoverable patch including existing uncommitted changes. Do not reset a worktree.
- Define a few observable acceptance checks from the user's goal: correct result,
  important constraints retained, unnecessary work removed. Do not ask the user to
  choose a generic checklist or equate checklist completion with quality.
- Prepare 3–5 representative tasks with normal and adjacent/confusing cases.
  Reserve at least one new case from influencing the edits. Keep the same inputs,