The Agent Skills Directory

[SAFE]: The skill is comprised entirely of Markdown files (SKILL.md, persona.md, and 23 topic files) containing instructional prose and marketing frameworks. It does not include any executable scripts, binary files, or configuration files that could trigger malicious behavior.- [PROMPT_INJECTION]: The instructions focus on routing logic and persona emulation. There are no attempts to override agent safety protocols, extract system prompts, or bypass platform constraints.- [DATA_EXFILTRATION]: No network operations (such as curl, wget, or fetch) are used to transmit data externally. The skill does not attempt to access sensitive system paths, credentials, or environment variables.- [REMOTE_CODE_EXECUTION]: The skill does not download or execute external code. There are no package manifest files (package.json, requirements.txt) and no installation commands present.- [INDIRECT_PROMPT_INJECTION]: The skill defines a surface for processing untrusted user data (e.g., reviewing user-provided copy), but it possesses no actionable capabilities (filesystem, network, or shell access) that could be exploited. The risk is limited to text generation behavior, which is consistent with the skill's purpose.- [COMMAND_EXECUTION]: No shell commands or system-level invocations were detected in any of the analyzed files, including the SKILL.md file (Category 11 check).

dojo-harry-dry