ClawHub
Back to skill

Security audit

Social Posting

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly coherent for social posting, but it allows public posts through global provider credentials when a user has no personal credentials, which needs review before use.

Install only if you are comfortable with a skill that can publish or schedule public social media posts. Before use, require a workflow that blocks posting unless the exact provider account, platform list, content, media, schedule, and credential source are shown and confirmed, and disable global credential fallback for ordinary user-generated posts unless it is intentionally a service-owned publishing flow.

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 (4)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger language includes broad phrases like 'social post' and 'any cross-platform publishing request,' which can cause the skill to activate on ambiguous requests outside the user's intent. In an agent environment, overbroad activation increases the chance of unintended execution of a high-impact action: publishing content to external social platforms. This context makes the issue more serious than a generic routing bug because accidental invocation can lead directly to unauthorized public posting.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 36)May include surrounding context.

md
| Provider | Role | API Base |
|----------|------|----------|
| **PostForMe** | Primary (cheaper) | `https://api.postforme.dev/v1` |
| **LATE** | Fallback (reliable) | `https://getlate.dev/api/v1` |

Failover order:

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill documents global credential fallback but does not prominently warn users that, absent personal credentials, their content may be published through shared provider accounts. This missing disclosure undermines informed consent and can result in users unintentionally posting under the wrong identity or to service-managed accounts. Because the skill performs public external actions, lack of account-selection transparency materially increases harm.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The documented failover explicitly allows posting with global environment-variable provider credentials when a user lacks personal credentials. In a multi-user social-posting skill, this can cause content to be published through a shared service account without explicit user consent or clear account attribution, creating cross-tenant misuse, accidental publication, and reputational damage. The social-posting context makes this more dangerous because the action is externally visible and irreversible once sent to third-party platforms.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.