The Agent Skills Directory

[SAFE]: The analyzed files consist exclusively of Markdown instructions and metadata for agent behavior. There are no patterns suggesting data exfiltration, credential theft, or unauthorized file access.
[NO_CODE]: No scripts, executables, or external software packages are distributed with this skill. It relies on the agent's native ability to parse text and generate ASCII-based visual representations.
[INDIRECT_PROMPT_INJECTION]: The skill processes inventories of decisions to build dependency graphs, creating a surface for indirect prompt injection. Untrusted data enters via decision lists processed at runtime. While explicit boundary markers are not specified, the skill possesses no capabilities for command execution, network access, or file modification, effectively neutralizing the threat of malicious instructions embedded in the processed data.

decision-path