ClawHub
Back to skill

Security audit

Archive

Security checks for vulnerabilities and agentic risk

Overview

This skill is a straightforward local archiving helper that stores user-selected content under ~/archive and can resurface relevant saved items, with no evidence of hidden execution or exfiltration.

Install this only if you want a persistent local archive of saved content. Be aware it may store full extracted text, source URLs, local file paths, search history, tags, project names, and resurfacing preferences under ~/archive; avoid archiving sensitive material unless you are comfortable with that local record.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (7)

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 · SKILL.md (reported line 85)May include surrounding context.

md
- Surface ONLY if genuinely relevant (max 1-2 per session)
- Include context: "You saved this 3 months ago when researching X"

### 6. Never Delete Without Asking
- Old items → mark as "possibly outdated", don't delete
- Duplicates → merge, keep both URLs
- Project closed → archive to cold storage, don't remove

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 · SKILL.md (reported line 107)May include surrounding context.

md
This skill NEVER:
- Monitors or observes without explicit request
- Deletes content without confirmation
- Modifies original archived content
- Accesses external services without user action

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The trigger phrases are broad, common expressions that can easily occur in normal conversation without clear archival intent. This creates a risk of unintended activation, causing the agent to fetch, extract, and store potentially sensitive user or external content when the user did not mean to invoke the skill.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill instructs fetching full web content and storing local PDF paths, but it does not warn users that external resources may be retrieved or that local file information may be preserved. In a capture/archive context, this increases privacy and data handling risk because users may unknowingly cause collection of sensitive remote content, metadata, or local path information.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The template explicitly stores user preferences, resurfacing behavior, search metadata, and usage-derived statistics in a local memory file without any indication of consent, minimization, retention limits, or privacy notice. Even if kept locally, this creates a persistent behavioral profile that can expose sensitive habits, projects, and interests if accessed by other local processes, backups, or unauthorized users.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The template instructs the system to archive complete extracted content and maintain a search/access log, which can accumulate copyrighted material, sensitive text, and a detailed record of user queries and engagement over time. Without disclosure, retention boundaries, or safeguards, this increases privacy and data-exposure risk and may preserve information far beyond what the user expects.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The activation conditions rely on broad phrases such as 'matches archived item' and 'related archives' without defining how close the match must be or giving concrete inclusion/exclusion examples. Although a later relevance threshold exists, these trigger lines themselves are vague enough to risk unintended resurfacing during ordinary conversation.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.