competitor-alternatives
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes competitor data and reviews which may include untrusted external content.
- Ingestion points: The "Deep Research Process" in
SKILL.mdinvolves gathering data from review platforms and competitor websites. This data is then processed by the agent and the included Python scripts. - Boundary markers: While the skill emphasizes "Honest positioning" and "Verifiable claims," it lacks explicit instructions for the agent to treat ingested data as untrusted or to use specific delimiters.
- Capability inventory: The provided scripts are restricted to reading local files and performing data processing using standard libraries. They do not perform network operations, subprocess calls, or dynamic code execution.
- Sanitization: There is no evidence of automated sanitization or filtering of the ingested external content.
- [COMMAND_EXECUTION]: The skill documentation includes examples for executing local Python scripts via the command line.
- The scripts are designed for data analysis and planning (e.g.,
python scripts/comparison_page_planner.py). - Technical review of the source code for all three scripts confirms they are safe, utilizing only standard Python libraries for file I/O and data manipulation without risk of command injection or unauthorized access.
Audit Metadata