brand-marketing-strategy-proposal-generator

Warn

Audited by Gen Agent Trust Hub on Jun 30, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSPROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The installation instructions direct users to clone a repository from https://github.com/yuleiwang156-a11y/brand-marketing-strategy-proposal-skill.git. This repository is hosted by a third-party account that does not align with the stated skill author (aradotso), posing a supply chain risk.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted external data as primary instructions for its workflow.
  • Ingestion points: The load_client_materials function and the 'Research & Analysis' workflow ingest content from user-provided files in inputs/客户资料/, including PDF, XLSX, and ZIP files.
  • Boundary markers: Absent. The instructions do not define delimiters or provide 'ignore instructions' warnings to the agent when parsing client documents.
  • Capability inventory: The skill can read local files, write structured documents using python-docx, and generate content based on input data.
  • Sanitization: Absent. Content extracted from client materials is directly interpolated into research reports and PPT scripts without validation or filtering.
  • [DATA_EXFILTRATION]: The skill processes highly sensitive business information, including client briefs and market data. If a malicious client file is processed (see Prompt Injection), the lack of security boundaries could be exploited to leak this sensitive information or influence the generated strategy proposals.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jun 30, 2026, 05:24 PM
Security Audit — agent-trust-hub — brand-marketing-strategy-proposal-generator