ClawHub
Back to skill

Security audit

Iso 27001 Internal Audit

Security checks for vulnerabilities and agentic risk

Overview

This is a markdown-only ISO 27001 audit guidance skill with expected evidence-collection commands, but users should protect any audit outputs they collect.

Before installing, treat this as procedural audit guidance rather than an automated scanner. Run commands only against systems you are authorized to audit, store evidence in approved encrypted locations, redact unnecessary personal or admin details, and avoid pasting raw exports or tokens into unsecured chats or reports.

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)

Vague Triggers

Medium
Confidence
91% confidence
Finding
The manifest description lists invocation phrases, and "control assessment" is generic enough to match many routine conversations outside this specific skill's intended ISO 27001 internal-audit context. While some other examples are specific, this phrase lacks narrowing constraints or exclusion examples that would reduce accidental activation.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The skill directs users to collect and store audit evidence and screenshots locally, but it does not warn that such evidence may contain sensitive security, personnel, system configuration, or customer data. This can lead to inadvertent accumulation of unprotected audit artifacts on disk, increasing exposure if endpoints are shared, compromised, or insufficiently encrypted.

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
### DON'T
- Screenshot portals without visible system clock — auditors will reject undated evidence
- Accept "we have a policy" without checking implementation — "show me" > "tell me"
- Audit your own work — independence requirement (Clause 9.2) means auditors can't audit their own area
- Treat checkbox controls as zero-effort — even N/A controls need justification in the SoA
- Skip ISMS clauses to focus only on Annex A — most first-time failures are in clauses 4-10
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.

External Transmission

Medium
Category
Data Exfiltration
Content
gh api repos/{owner}/{repo}/security-advisories --paginate | jq '.[0:5]'

# If using PagerDuty:
# curl -H "Authorization: Token token={api_key}" https://api.pagerduty.com/incidents?limit=10
```

---
Confidence
50% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

External Transmission

Medium
Category
Data Exfiltration
Content
gh api repos/{owner}/{repo}/security-advisories --paginate | jq '.[0:5]'

# If using PagerDuty:
# curl -H "Authorization: Token token={api_key}" https://api.pagerduty.com/incidents?limit=10
```

---
Confidence
50% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Missing User Warnings

Low
Confidence
91% confidence
Finding
The skill instructs auditors to run commands that enumerate organization members, IAM bindings, role assignments, and directory users, which can expose sensitive identity and authorization data. In an audit context this collection is expected, but the lack of an explicit warning about handling, minimizing, and protecting the resulting output creates a real data-exposure risk if evidence is copied into chats, tickets, or unsecured files.

Missing User Warnings

Low
Confidence
94% confidence
Finding
The privileged-access section includes queries for high-value administrative data, including cloud owner/editor roles, GitHub admins, and Microsoft Graph Global Administrator membership. Although this is legitimate for an internal audit, the commands surface especially sensitive administrative relationships, and the file provides no warning that this information should be tightly controlled and not broadly disclosed.

Static analysis

No suspicious patterns detected.