pp-clickhouse-cloud

Pass

Audited by Gen Agent Trust Hub on Jul 28, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches the Clickhouse Cloud CLI and MCP server from the author's GitHub repository and npm registry during setup.
  • Evidence: npx -y @mvanhorn/printing-press-library install clickhouse-cloud --cli-only in SKILL.md.
  • Evidence: go install github.com/mvanhorn/printing-press-library/library/cloud/clickhouse-cloud/cmd/clickhouse-cloud-pp-cli@latest in SKILL.md.
  • [DATA_EXFILTRATION]: The CLI includes a --deliver flag that can route sensitive command output, such as service details or organization configurations, to external URLs via webhooks or local file paths.
  • Evidence: Documentation for the --deliver flag in SKILL.md supports webhook:<url> and file:<path> sinks.
  • [COMMAND_EXECUTION]: The skill executes shell commands using the Bash tool to perform installation, authentication, and management tasks.
  • Evidence: Frequent usage of clickhouse-cloud-pp-cli, go install, and npx commands within the instructions.
  • [PROMPT_INJECTION]: The skill demonstrates an indirect prompt injection surface by ingesting untrusted data from the ClickHouse Cloud API and feedback mechanisms.
  • Ingestion points: clickhouse-cloud-pp-cli feedback and command outputs from various subcommands (e.g., services, members, organizations).
  • Boundary markers: The instructions lack explicit delimiters or warnings to ignore embedded instructions in the processed data.
  • Capability inventory: The skill has access to the Bash tool, file system writes, and network transmission via the --deliver flag.
  • Sanitization: No sanitization or validation of data retrieved from remote endpoints is documented.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 28, 2026, 09:18 AM
Security Audit — agent-trust-hub — pp-clickhouse-cloud