The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/move_cards.py executes the rumdl utility using subprocess.run to perform link validation. The command is constructed as a list, which prevents shell injection vulnerabilities.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it reads and processes markdown cards provided by the user. * Ingestion points: Card files are read from the backlog/, active/, and archive/ directories by scripts/move_cards.py and scripts/validate_frontmatter.py. * Boundary markers: There are no explicit delimiters or instructions to the agent to ignore content-embedded commands within the cards. * Capability inventory: The skill has the ability to move and delete files within the board structure and execute external processes. * Sanitization: The skill correctly uses yaml.safe_load in scripts/shared.py to prevent code execution during the parsing of card frontmatter.

mols-kanban-markdown