The Agent Skills Directory

[SAFE]: The skill is purely instructional and does not contain malicious code, hidden commands, or obfuscated logic. It is designed to assist an agent in performing layout calculations and SVG generation based on provided source material. No malicious patterns or high-risk command execution were detected.
[DATA_EXPOSURE]: While the skill reads local source files and writes SVG/Markdown outputs to the filesystem, these operations are restricted to the intended functional scope of diagram generation and project documentation. No evidence of unauthorized data access, credential harvesting, or exfiltration to external domains was found.
[REMOTE_CODE_EXECUTION]: The skill does not perform any network operations, download external dependencies, or utilize dynamic execution environments that could be exploited for remote code execution. It relies on the agent's internal capabilities to write SVG code following the provided reference documentation.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-provided source material (text or files). While this represents a data ingestion surface, the risk is minimized by the structured workflow requiring a manual confirmation of the diagram plan before generation. The capability is restricted to writing specialized SVG and Markdown files, which limits the potential impact of instructions embedded in the processed data.

baoyu-diagram