ClawHub
Back to skill

Security audit

Loop Engineering

Security checks for vulnerabilities and agentic risk

Overview

This is a documentation-only skill about designing agent loops, with some broad examples but no hidden execution or persistence.

Install this as educational guidance for designing loop systems. Review the scheduling and auto-fix examples before adapting them, and require explicit approvals for deployment, dependency changes, broad write access, or long-running automated loops.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (4)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill's trigger list includes very broad terms such as "system architect" and general AI-programming phrases, which can cause the skill to activate for requests outside its narrowly intended topic. In agentic environments, overbroad activation can misroute user intent, inject irrelevant guidance into unrelated tasks, and increase the chance that autonomous workflows apply loop-oriented advice where it is unsafe or inappropriate.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The example trigger words for loop-specific skills are generic enough to match ordinary user requests, which can cause unintended auto-activation of privileged or side-effecting skills. In a platform where skills load automatically based on content routing, ambiguous triggers increase the risk of incorrect tool selection, unexpected actions, and prompt-surface expansion across unrelated tasks.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
78% confidence
Finding

The natural-language comment 'Ensure model differs from executing Agent' is paired with a hard-coded model value 'claude-sonnet', which prescribes a specific model choice rather than offering an option. Under the stated policy category, fixed natural-language constraints on model/language-style choices can be a policy concern when not presented as optional or justified.

Content

No source excerpt is available for this finding.

Scope Creep

Low
Category
Excessive Agency
Confidence
75% confidence
Finding

Skill's behavior or capabilities extend beyond its stated purpose. Scope creep allows an agent to perform actions unrelated to its documented functionality, increasing the attack surface.

Content

Scanner excerpt · references/integration.md (reported line 231)May include surrounding context.

text
Phase 1: Select 1-2 tasks suitable for looping as pilots
Phase 2: After validation and stabilization, gradually expand scope
Phase 3: Establish team-internal loop design specifications and templates
Phase 4: Distill mature loop patterns into reusable Skills