The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local Python scripts (scaffold_bundle.py, okf_linter.py, index_generator.py) to manage business documentation. These tools are provided within the skill package and rely exclusively on the Python standard library for file operations and text processing without involving dangerous system commands.
[DATA_EXFILTRATION]: No network activity was detected. The skill only reads and writes markdown files in a user-specified local directory related to the business documentation project.
[PROMPT_INJECTION]: The instructions follow a professional, structured interview format. There are no attempts to bypass safety filters, extract system prompts, or override agent constraints.
[INDIRECT_PROMPT_INJECTION]: The skill ingests user input and potentially web search data (Phase 3). While this creates an injection surface, the risk is minimized by the skill's specific focus on generating static markdown documentation.
Ingestion points: User interview responses and web search results (Phase 3).
Boundary markers: Uses markdown structure and specific tags like [SUPOSIÇÃO], though it lacks rigorous isolation for external content.
Capability inventory: Local file read/write and execution of provided Python scripts.
Sanitization: File names are sanitized via a slugify function, and content is processed by the AI model for documentation purposes.

arquiteto-de-empresa