The Agent Skills Directory

[DATA_EXPOSURE]: The skill reads from and writes to a configuration file located at ~/plotloom.toml. While this is part of the tool's registry management, it involves accessing the user's home directory context.
[COMMAND_EXECUTION]: The skill instructions include the execution of the plotloom CLI for repository initialization and asset generation, which involves spawning subprocesses with arguments derived from user input.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data from user ideas and existing markdown files (series.md, characters.md) without explicit sanitization or boundary markers.
Ingestion points: User input, existing series and character markdown files, and the ~/plotloom.toml registry.
Boundary markers: None identified in instructions or templates.
Capability inventory: The skill has the ability to write project files and execute local CLI commands via plotloom.
Sanitization: No explicit logic is provided to sanitize or validate external data before it is interpolated into files or command arguments.
[DYNAMIC_EXECUTION]: The skill workflow involves generating textual briefs and prompt files which are subsequently passed to a CLI tool for execution, representing a dynamic content generation and execution path.

plotloom-series-brainstorming