ClawHub
Back to skill

Security audit

Evidence-First Research

Security checks for vulnerabilities and agentic risk

Overview

This skill is a plain research-workflow guide that tells the agent to check evidence before making scientific or medical claims.

Install this if you want Codex to slow down and verify scientific or medical claims against stronger sources before drafting plans or recommendations. Expect it to encourage searches and evidence-quality caveats; verify current medical or regulatory guidance for high-stakes decisions.

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 (2)

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
## What It Tries To Prevent

- starting analysis as if the problem were novel without checking the literature
- reinventing methods that already have strong precedents
- overclaiming from weak, indirect, or mismatched evidence
- skipping safety, guideline, or reproducibility checks in biomedical work
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.

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
## What It Tries To Prevent

- starting analysis as if the problem were novel without checking the literature
- reinventing methods that already have strong precedents
- overclaiming from weak, indirect, or mismatched evidence
- skipping safety, guideline, or reproducibility checks in biomedical work
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.

Static analysis

No suspicious patterns detected.