pp-docker-hub
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches and installs the
docker-hub-pp-clianddocker-hub-pp-mcpbinaries from the vendor's GitHub repository (github.com/mvanhorn/printing-press-library) and from the NPM registry via the@mvanhorn/printing-press-librarypackage. - [REMOTE_CODE_EXECUTION]: Utilizes
go installandnpxto download and execute code from remote sources during the prerequisite installation and MCP registration processes. - [COMMAND_EXECUTION]: Executes shell commands to install tools, verify system paths, check binary versions, and perform Docker Hub API queries through the CLI.
- [DATA_EXFILTRATION]: The CLI includes a
--deliver webhook:<url>feature that POSTs command output to an external URL. It also supports sending feedback to a remote endpoint defined by theDOCKER_HUB_FEEDBACK_ENDPOINTenvironment variable. - [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the Docker Hub public API, which represents a surface for indirect prompt injection. 1. Ingestion points: Docker Hub search results and repository metadata processed by the agent via
SKILL.md. 2. Boundary markers: Absent. 3. Capability inventory: The skill has file-write and network-send capabilities via the CLI's--deliverflag. 4. Sanitization: No sanitization of API data is specified.
Audit Metadata