pp-daraz
Pass
Audited by Gen Agent Trust Hub on Jul 28, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the agent to install a CLI tool using
npx -y @mvanhorn/printing-press-libraryandgo install github.com/mvanhorn/printing-press-library. These downloads originate from the author's public repositories. - [DATA_EXFILTRATION]: The
daraz-pp-clitool includes a--deliver webhook:<url>parameter that can POST command output to an external server. Additionally, thefeedbackcommand can be configured to send local data to a remote endpoint via theDARAZ_FEEDBACK_ENDPOINTenvironment variable. - [COMMAND_EXECUTION]: The tool allows writing output directly to the file system using the
--deliver file:<path>option, which could be used to overwrite sensitive files if the path is not strictly controlled. - [PROMPT_INJECTION]: The skill processes untrusted product data and reviews from Daraz.pk, creating a surface for indirect prompt injection.
- Ingestion points: Product listings, descriptions, and reviews fetched from Daraz.pk via the
productsandreviewscommands. - Boundary markers: None identified in the skill instructions to separate external data from agent instructions.
- Capability inventory: The skill uses a tool capable of network requests (webhooks), file writing, and bash command execution.
- Sanitization: No specific sanitization or validation of the external marketplace data is documented before it is passed to the agent context.
Audit Metadata