customer-feedback-analysis
Pass
Audited by Gen Agent Trust Hub on Oct 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local helper script
scripts/crawlora.shto perform API requests. The script implements security best practices, including parameter sanitization, whitelisting allowed API routes, and preventing sensitive credentials from appearing in process lists by using temporary configuration files. - [EXTERNAL_DOWNLOADS]: Fetches customer feedback data from the vendor's official API at
api.crawlora.netto perform its primary function of sentiment and theme analysis. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted customer reviews and social media comments harvested from third-party platforms like Reddit and various app stores.
- Ingestion points: API responses containing raw text from reviews and social media comments processed in
SKILL.md. - Boundary markers: The skill provides structural guidance for analysis (e.g., requesting a theme table and short attributed examples), but lacks explicit technical delimiters to isolate untrusted review text within the agent's context.
- Capability inventory: The skill has the ability to execute the local
scripts/crawlora.shscript to fetch data. - Sanitization: While the helper script sanitizes request parameters, there is no evidence of sanitization or filtering applied to the content of the reviews themselves before they are interpreted by the agent.
Audit Metadata