infra-postgres
Pass
Audited by Gen Agent Trust Hub on Nov 2, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches the
clickhousectlCLI tool directly from the vendor's official domain (clickhouse.com). This is a standard installation procedure for the vendor's tooling. - [REMOTE_CODE_EXECUTION]: A installation script is downloaded and executed via
curl -fsSL https://clickhouse.com/cli | sh. This operation targets the official vendor infrastructure to install the necessary management utility. - [COMMAND_EXECUTION]: The skill makes extensive use of local shell commands to manage database infrastructure, including:
dockercommands to manage local Postgres containers.psql,pg_dump, andpg_restorefor database interaction and migration.- The
clickhousectlCLI for cloud service management. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to process output from various external sources, creating a surface for indirect instructions.
- Ingestion points: Instructions in
SKILL.md,ref/local.md,ref/cloud.md, andref/migrate.mdinvolve reading and acting upon JSON output fromclickhousectl, query results frompsql, and logs fromdockerorpg_dump. - Boundary markers: The instructions do not define explicit boundary markers or "ignore instructions" warnings for the data ingested from these tools.
- Capability inventory: The skill has significant capabilities, including executing shell commands, performing file system operations (writing to
.env,ca.pem, anddump/directories), and conducting network operations through the vendor's CLI. - Sanitization: There is no explicit sanitization or escaping of the external data before it is presented to or processed by the agent.
Audit Metadata