The Agent Skills Directory

[SAFE]: The skill is focused on narrative structure (Goal-Conflict-Disaster) and does not contain any malicious instructions or patterns.
[COMMAND_EXECUTION]: The skill includes a local utility scripts/analyze-scene.ts. This script uses Deno with restricted permissions (--allow-read) to perform text-based analysis via regular expressions. It does not perform network operations, access sensitive credentials, or execute code based on user input.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-provided scene text. The processing is limited to keyword counting and basic heuristics. The results are returned as descriptive markdown reports, presenting a negligible security risk.

scene-sequencing