The Agent Skills Directory

[Indirect Prompt Injection] (MEDIUM): The skill possesses a surface for indirect prompt injection as it ingests untrusted data from external files and writes to the file system. * Ingestion points: Untrusted data enters the context via load(), fill_from_json(), and batch_fill() (referenced in SKILL.md). * Boundary markers: Absent; there are no delimiters or instructions to ignore embedded commands in the processed data. * Capability inventory: The skill has file-write capabilities via save() and batch_fill(). * Sanitization: Absent; no validation or escaping of input data is documented.
[Missing Implementation] (MEDIUM): The documentation in SKILL.md refers to a form_filler.py script and a FormFiller class, but these files are missing from the skill directory. This prevents verification of potential Category 4 (Remote Code Execution) or Category 10 (Dynamic Execution) risks within the actual Python logic.

form-filler