The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the run_command tool and the vectorcraft-cli executable to perform graphic editing tasks such as creating graphs, setting types, and modifying data.
[EXTERNAL_DOWNLOADS]: The skill includes metadata referencing a portable Windows runtime for the Vectorcraft tool hosted on a GitHub repository (references/craft-command-evidence.json). Additionally, it provides instructions for installing a related skill using npx from the same author's namespace (full-aigc-skills/vectorcraft-skills).
[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied "material" to generate reports and charts, which presents a surface for indirect prompt injection attacks.
Ingestion points: The skill reads external material to determine delivery use, dimensions, and modification ranges as described in SKILL.md.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands within the ingested materials are provided.
Capability inventory: The skill has access to run_command, draw_shape, and save_file tools across its scripts, allowing for file system modifications and CLI interactions.
Sanitization: The skill lacks specified sanitization or validation logic for the content extracted from external materials before passing them to the rendering engine.

vectorcraft-editable-chart