ClawHub
Back to skill

Security audit

Programmatic Ad Analyst

Security checks for vulnerabilities and agentic risk

Overview

This is a markdown-only advertising analysis skill whose requested web search access fits its stated campaign-diagnosis purpose.

Reasonable to install for programmatic ad analysis. Treat it as advisory: verify current benchmark claims when making spend decisions, and provide only the campaign data needed for the analysis.

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
  • YARA SignaturesMalware Match, Webshell Match, Cryptominer Match
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

YARA rule 'info_stealer': Information stealer patterns (credential harvesting, browser data theft) [malware]

High
Category
YARA Match
Confidence
75% confidence
Finding

YARA rule matched a known malware signature (reverse shell, backdoor, ransomware, C2 framework, or info stealer).

Content

Scanner excerpt · SKILL.md (reported line 116)May include surrounding context.

md
segments | Medium | High |
| Contextual | Page content/URL | Medium | High |
| Lookalike | Model-based expansion | Medium | High |
| Behavioral | Cross-site history | Medium–Low | High |

**Post-cookie targeting stack (2025+):**
- **UID2 / RampID**: Hashed email-based identity, requires user consent
- **Google Privacy Sandbox / Topics API**: Interest cohort-based, replaces
  third-party cookies in Chrome, limited granularity
- **Publisher Provided IDs (PPID)**: Publisher-owned, highest match rate
  within that publisher's inventory
- **Contextual + first-party**: Most durable long-term approach

### Frequency Cap Diagnosis

Cookie-based frequency caps **fail silently** for iOS Safari (ITP),
Firefox (ETP), and private/incognito users. Your reported frequency is
likely understated. Signs of hidden overexposure:
- CTR declining week-over-week without budget changes
- Increasing CPA despite stable targeting

**Recommended frequency by objective:**

| Objective | Ca

Static analysis

No suspicious patterns detected.