add-marketplace

Pass

Audited by Gen Agent Trust Hub on Nov 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute various shell commands including git, npm, node, and npx for repository management, installation of dependencies, and validation of plugin configurations.
  • [DYNAMIC_EXECUTION]: The utility script scripts/check-hooks.sh extracts a command string from a project configuration file (hooks/hooks.json) and executes it using bash -c. This pattern allows for the dynamic execution of project-defined hook commands during the validation phase.
  • [EXTERNAL_DOWNLOADS]: The skill utilizes external packages and tools such as @postman/postman-plugin, @anthropic-ai/claude-code, and ajv. These resources are fetched from the author's own organization or well-known trusted entities. Additionally, the skill includes instructions for fetching official vendor schemas from remote URLs for CI validation.
  • [INDIRECT_PROMPT_INJECTION]: The skill involves the ingestion and processing of external vendor manifests and schemas. This creates a potential surface for indirect prompt injection if the remote resources are compromised, although the skill mitigates this by using schema validation tools like ajv to verify the structure of the data.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 17, 2026, 04:03 PM