ClawHub
Back to skill

Security audit

SocialKit YouTube Research

Security checks for vulnerabilities and agentic risk

Overview

The skill performs user-directed YouTube research with proportionate API access and local report creation.

Use a Google API key restricted to YouTube Data API v3; research does not require channel OAuth or upload permissions. Search terms go to Google, requests consume quota, and the skill creates a project report and local quota counters. All three bundled files were inspected; separately installed socialkit-social-core components were not assessed.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 19)May include surrounding context.

md
`node scripts/research.mjs --query "<terms>" --max 15 --order viewCount`

Skill declares no tool scope ('permissions' or 'allowed-tools') but code capabilities were detected: env, network.

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding

Without declared permissions the skill's intent is opaque and cannot be validated.

Content

No source excerpt is available for this finding.

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 22)May include surrounding context.

md
`node scripts/research.mjs --query "<terms>" --max 15 --order viewCount`
   Add `--published-after <ISO date>` for recent trends. Output is JSON on stdout.
   Each query costs one call from the 100/day search bucket; do not loop over many
   queries without asking.
3. Analyse: recurring angles, title patterns, typical length, views per day since
   publishing, and what the top videos do not cover.
4. Write `research.md` in the project folder with: queries used, a table of the top

The research client's implementation includes video-upload and thumbnail-modification capabilities unrelated to topic research.

Medium
Category
Excessive Agency
Confidence
96% confidence
Finding

The client implements resumable video uploads and thumbnail updates, including reading local media files and sending POST/PUT requests to YouTube. These remote-write capabilities are unnecessary for the manifest's research purpose; although the current CLI only invokes search and video listing with an API key, the bundled client still implements these additional operations.

Content

No source excerpt is available for this finding.