blox-cli
Pass
Audited by Gen Agent Trust Hub on Aug 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes shell commands using
node:child_processinscripts/sync-registries.mjs. Specifically, it callsgit cloneandgit pullto synchronize remote component registries. This is a standard operation for the skill's intended purpose of managing a software registry. - [EXTERNAL_DOWNLOADS]: The skill downloads registry metadata from
https://tfs.aas.com.sa/Medad/BLOX/_git/default-registry. This domain belongs to the vendor (Naseej), matching the author context provided. These downloads are part of the core registry synchronization feature. - [EXTERNAL_DOWNLOADS]: The script
scripts/local-config.mjsimports theenv-pathspackage. This is an external Node.js dependency required for the script to function, though it is not bundled within the provided skill files. - [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection by ingesting data from external registries and presenting it to the agent.
- Ingestion points:
scripts/get-blocks-context.mjsreadsindex.jsonfiles from synchronized registries and the.bloxconfiguration from the local project. - Boundary markers: None present. The agent is instructed to parse the JSON output directly and present descriptions to the user.
- Capability inventory: The skill can execute
nodescripts andgitcommands via subprocesses, and write to the local file system in standard config/data paths. - Sanitization: No explicit sanitization of text content (descriptions, names) retrieved from the registry before it is processed by the agent.
Audit Metadata