ClawHub
Back to skill

Security audit

LinkedIn Data

Security checks for vulnerabilities and agentic risk

Overview

This LinkedIn data skill mostly matches its stated purpose, but it also steers users toward dynamically expanded Superagnt platform tools beyond LinkedIn, which needs review before installation.

Install only if you are comfortable sending LinkedIn research queries and targets to Superagnt. Prefer using only the documented LinkedIn endpoints unless you intentionally want broader Superagnt platform access, and review any MCP tool expansion before enabling additional tool families.

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
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (7)

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill description does not clearly warn that user-supplied LinkedIn profile, company, job, and post queries are sent to an external third-party API service. This can mislead users about where potentially sensitive professional research targets and usage patterns are being transmitted, undermining informed consent and privacy expectations.

Context-Inappropriate Capability

Medium
Confidence
96% confidence
Finding
The document directs users to connect to an MCP server that can expose a growing tool surface and refresh tools dynamically. That creates a capability-expansion risk: a skill advertised as a fixed LinkedIn API wrapper may instead become a gateway to new tools and actions without the same static review or narrow purpose constraints.

External Transmission

Medium
Category
Data Exfiltration
Content
One call proves the key, the credit balance, and this source end to end:

```bash
curl -s https://api.superagnt.com/v1/credits \
  -H 'Authorization: Bearer $SUPERAGNT_API_KEY'
```
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
One call proves the key, the credit balance, and this source end to end:

```bash
curl -s https://api.superagnt.com/v1/credits \
  -H 'Authorization: Bearer $SUPERAGNT_API_KEY'
```
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
One call proves the key, the credit balance, and this source end to end:

```bash
curl -s https://api.superagnt.com/v1/credits \
  -H 'Authorization: Bearer $SUPERAGNT_API_KEY'
```
Confidence
50% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Description-Behavior Mismatch

Medium
Confidence
97% confidence
Finding
The skill is scoped as a LinkedIn-data integration, but it explicitly advertises broader platform capabilities including workspace database, files, webhooks, queues, and first-party enrichment. That scope expansion can cause an agent or operator to use capabilities unrelated to the declared purpose, increasing the chance of unintended data access or exfiltration beyond the user's expectation.

External Transmission

Medium
Category
Data Exfiltration
Content
is never charged without a human confirming in the dashboard.

```bash
curl https://api.superagnt.com/v1/platforms
```

## Links
Confidence
88% confidence
Finding
This call is not to the LinkedIn-data endpoint but to a broader platform-discovery endpoint, reinforcing that the skill can steer users beyond its declared narrow purpose. In context with the surrounding text about enabling more tool families, it increases the risk of capability creep and unintended discovery of unrelated services through the same credential.

Static analysis

No suspicious patterns detected.