roi-calculator
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local Python scripts (
campaign-tracker.pyandroi-calculator.py) using the system's Python interpreter. These scripts are invoked with parameters, such as the brand 'slug', which are derived from local configuration files. - [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from multiple sources and uses it to drive agent analysis and reporting.
- Ingestion points: The skill reads user-provided campaign metrics (spend, revenue, conversions) and brand configuration files located in
~/.claude-marketing/brands/. - Boundary markers: There are no explicit delimiters or safety instructions specified to isolate untrusted input data from the execution instructions.
- Capability inventory: The skill possesses the ability to execute Python scripts via subprocess calls and perform read/write operations on the local filesystem.
- Sanitization: The process does not include explicit validation or sanitization steps for the data before it is passed to the analysis tools or incorporated into the final report.
Audit Metadata