The Agent Skills Directory

[PROMPT_INJECTION]: The skill's instructions are focused entirely on providing domain-specific advice for Contentstack. There are no attempts to override agent behavior, bypass safety filters, or extract system prompts.
[DATA_EXPOSURE]: The skill includes explicit security sections mandating the use of environment variables for API keys and tokens. It prohibits the hardcoding of secrets in client-side code and ensures that management tokens are never revealed.
[REMOTE_CODE_EXECUTION]: The skill is defined as advisory-only and restricted to read-only tools (Read, Grep, Glob). It does not include any commands for downloading external scripts, installing packages, or executing remote code.
[INDIRECT_PROMPT_INJECTION]: While the skill can ingest file content via the Read tool, it has no high-risk capabilities (like network operations or code execution) that could be exploited by malicious data in the processed files. It also lacks write capabilities, minimizing the surface for multi-step injection attacks.
[COMMAND_EXECUTION]: No dangerous system commands, shell escapes, or privilege escalation attempts were found. The skill does not use sudo, chmod, or any persistence mechanisms.

cms-variants-personalization