keyword-research
Warn
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: MEDIUMCREDENTIALS_UNSAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [CREDENTIALS_UNSAFE]: The skill accesses sensitive information by reading brand profiles and configuration files stored in the user's home directory at
~/.claude-marketing/brands/. These files contain brand-specific metadata, guidelines, and context necessary for the skill's operations. - [COMMAND_EXECUTION]: The skill executes a local Python script (
campaign-tracker.py) using a shell command to retrieve and list campaign history for the active brand. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources, including user-provided URLs and third-party SEO tool outputs (Ahrefs, Semrush, SE Ranking, and Google Search Console), without explicit sanitization or boundary markers.
- Ingestion points: User-provided URLs for theme extraction and results from connected SEO MCP tools.
- Boundary markers: None identified; instructions do not specify the use of delimiters or warnings to ignore embedded instructions in external data.
- Capability inventory: Execution of local scripts and reading of sensitive local configuration files.
- Sanitization: No evidence of content escaping, validation, or filtering of external data before processing.
Audit Metadata