Back to skill

Security audit

facebook-data

Security checks for vulnerabilities and agentic risk

Overview

ClawScan marks this skill benign with high confidence; the scanners report documented Facebook data access alongside a few capabilities or requests that may not fully match its stated purpose.

The verdict is benign, not a guarantee of zero risk. Before installing, consider whether you need the people-search feature and are comfortable with the documented Superagnt API use; the credit-check request is the least clearly related capability. Skillspector reports complete inspection of the two listed files, while code-focused checks were not applicable because no files matched their criteria.

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
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (6)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
90% confidence
Finding

The supplied chunk only requests account credits, which is materially different from the advertised Facebook functionality. A credit check could support a larger integration, but this chunk shows no connection to a Facebook operation that would establish it as a supporting implementation detail.

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 64)May include surrounding context.

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'

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 1075)May include surrounding context.

Example

bash
curl -X GET 'https://api.superagnt.com/v1/data/facebook&; \
  -H 'Authorization: Bearer $SUPERAGNT_API_KEY' \
  -H 'Content-Type: application/json'

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 74)May include surrounding context.

Base URL

text
https://api.superagnt.com/v1/data/facebook

Available Endpoints

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 1094)May include surrounding context.

md
nothing else. The same API key also works with superagnt's other data sources
and platform tools, but those are separate listings that the user installs or
enables themselves; this skill does not add or enable anything beyond what is
documented here. The public catalog is at `https://api.superagnt.com/v1/platforms`.

## Links

The skill exposes people search beyond the Facebook surfaces named in its manifest.

Medium
Category
Excessive Agency
Confidence
88% confidence
Finding

The manifest at L0003 describes access to pages, posts, groups, Marketplace listings, and the Ad Library. L0982–L1005 additionally expose a standalone people-search operation with query, location, and pagination parameters, enabling discovery of individuals rather than retrieval of the declared content surfaces.

Content

No source excerpt is available for this finding.