datocms-cli
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill inherently processes untrusted data from DatoCMS projects (schemas, records, and project metadata) which enters the agent's context through tools like
schema:inspectandcma:call. - Ingestion points: Data is ingested from the DatoCMS Content Management API via
npx datocms schema:inspectandnpx datocms cma:callas described inreferences/schema-inspect.mdandreferences/direct-cma-calls.md. - Boundary markers: No explicit boundary markers or "ignore embedded instructions" warnings are defined for the data returned by these tools.
- Capability inventory: The skill can execute shell commands via
npx, write files to the repository (e.g., migrations inreferences/creating-migrations.md, generated types inreferences/schema-generate.md), and perform network operations against the DatoCMS API. - Sanitization: The provided helper scripts
scripts/datocms-release.mjsandscripts/datocms-sync-projects.mjsinclude sanitization by validating environment IDs against a strict alphanumeric regex (/^[a-z0-9-]+$/). - [DYNAMIC_EXECUTION]: The skill facilitates the execution of dynamic scripts through
npx datocms cma:scriptandnpx datocms migrations:run. This allows for ad-hoc TypeScript operations and automated schema migrations. These features are documented with safety warnings regarding schema mutations and instructions to target sandbox environments for testing. - [EXTERNAL_DOWNLOADS]: The skill documents the installation of DatoCMS CLI plugins using
npx datocms plugins:add. Examples provided target official vendor resources such as@datocms/cli-plugin-wordpressand@datocms/cli-plugin-contentful. - [COMMAND_EXECUTION]: The skill includes Node.js automation scripts (
scripts/datocms-release.mjsandscripts/datocms-sync-projects.mjs) that usenode:child_processspawnSyncto invoke the DatoCMS CLI. These scripts are authored using security best practices, such as avoiding the logging of sensitive command arguments and using environment variables for credential management.
Audit Metadata