walmart-research
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The
scripts/crawlora.shhelper script usescurlto interact with the Crawlora API with several security protections. It stores the API key in a temporary file with owner-only permissions and ensures its deletion on exit, preventing exposure in process lists. It also restricts network requests to a specific whitelist of Walmart-related routes and prevents local file disclosure by rejecting sensitive characters in input parameters. - [DATA_EXPOSURE]: The skill manages the
CRAWLORA_API_KEYthrough environment variables, which is a standard and safe method for handling credentials. The helper script ensures the key is never passed as a command-line argument. - [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external Walmart pages through the Crawlora API. This surface is mitigated by the following factors:
- Ingestion points: Data enters through the
/walmart/searchand/walmart/productendpoints (found inreference/endpoints.md). - Boundary markers: None explicitly defined in the prompt instructions.
- Capability inventory: The skill possesses network request capabilities via
scripts/crawlora.shto a fixed base URL but lacks file system write access or arbitrary code execution capabilities. - Sanitization: The API returns normalized JSON rather than raw HTML, which reduces the complexity of processed content.
Audit Metadata