Content
88%Weight 40%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
A highly actionable, well-sequenced CLI skill body with strong validation checkpoints and feedback loops around destructive and batch operations. The main improvement opportunity is splitting the lengthy command reference and workflow detail into one-level-deep reference files to improve progressive disclosure.
Suggestions
Move the Command Groups table and the per-command `tool show`/`tool run` examples into a dedicated REFERENCE.md (one level deep) so SKILL.md stays a concise overview.
Tighten the autonomous-task broadcast paragraph in 'Manage Autonomous Agent Tasks' to the essential behavioral rule, trimming the explanatory prose.
Consider extracting the Season Validation episode-range logic into a separate reference file, since it is the most procedurally dense section and only needed for TV-season requests.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean and operational, dominated by concrete commands with only modest domain-specific prose (e.g., the autonomous-task broadcast behavior); a few explanatory sentences could still be trimmed, so it sits just below the fully lean anchor. | 4 / 5 |
Actionability | Provides copy-paste-ready `moviepilot tool run ...` commands with real parameter examples (e.g., `search_torrents media_source="themoviedb" media_id=791373 sites='1,3'`) covering the common cases across every command group. | 5 / 5 |
Workflow Clarity | Multi-step workflows are explicitly numbered with validation checkpoints ("Stop — do not call get_search_results yet", the mandatory Check Library and Subscriptions step, "confirm with user first — irreversible" for deletions) and an error-handling table that gives fix-and-retry feedback. | 5 / 5 |
Progressive Disclosure | No bundle files exist and the skill is a single well-sectioned document (Scope, Discover Commands, Command Groups, Workflows, Error handling); structure is good, but at ~285 lines the command-group reference and per-workflow detail could plausibly be split into one-level-deep reference files. | 4 / 5 |
Total | 18 / 20 Passed |