share-of-voice
Pass
Audited by Gen Agent Trust Hub on Apr 1, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill invokes local utility scripts
competitor-tracker.pyandgeo-tracker.pyto handle data persistence and competitive visibility auditing. These executions are core functional components of the skill's analytics workflow. - [DATA_EXPOSURE]: The process involves reading from and writing to configuration files located in the
~/.claude-marketing/directory, including brand profiles and competitor baselines. These operations are restricted to the application's specific workspace. - [INDIRECT_PROMPT_INJECTION]: The skill has a potential exposure surface for indirect prompt injection.
- Ingestion points: Social media mentions (Brandwatch MCP), search ranking data, and advertisement insights (SKILL.md).
- Boundary markers: None specified in the instructions to delimit external content from processing logic.
- Capability inventory: Execution of
competitor-tracker.pyand file operations within the~/.claude-marketing/directory (SKILL.md). - Sanitization: No explicit sanitization or escaping of ingested social media content is mentioned prior to sentiment analysis.
Audit Metadata