ClawHub
Back to skill

Security audit

AI Prompt Researcher

Security checks for vulnerabilities and agentic risk

Overview

This skill is a coherent GEO prompt research helper that only generates local guidance or report files, with no hidden network, credential, or persistence behavior found.

Before installing, treat it as a local report generator: provide only business context you are comfortable putting into a report, and specify output filenames carefully because the script can overwrite the chosen destination. The advanced usage examples reference scripts that are not included in the artifact, so rely on the included research_prompts.py workflow unless you verify additional files separately.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
88% confidence
Finding
The skill advertises executable workflows that write output files via commands like `python scripts/research_prompts.py --output report.md`, but it declares no `permissions` or `allowed-tools` scope. That mismatch can let an agent invoke file-writing behavior without an explicit least-privilege boundary, increasing the risk of unintended file creation or overwriting when the skill is auto-selected.

Vague Triggers

Medium
Confidence
95% confidence
Finding
The description contains broad trigger phrases such as 'Use whenever the user mentions researching AI search queries' and similar generic research wording, which can cause the skill to match many ordinary requests beyond its narrow intended purpose. Over-broad invocation increases the chance of unnecessary execution of a skill with code/file-write behavior, potentially causing unintended actions or data handling in unrelated contexts.

Static analysis

No suspicious patterns detected.