modules/programs/agents/shared/skills/radarr/SKILL.md
Search, add, inspect, and manage movies in my self-hosted Radarr library. Use when the user asks about movies, mentions Radarr, asks to add/remove a film, check what's downloading, what's upcoming, what's missing, view queue/history, manage collections, or trigger searches.
npx skillsauth add MichaelVessia/nixos-config radarrInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Manage my self-hosted Radarr (v3 API) movie library: search and add movies, inspect the library, see what's downloading/upcoming/missing, trigger searches, add full collections, remove movies.
Credentials are exported into the shell by sops-nix (see
modules/programs/shell.nix):
RADARR_URL — base URL (no trailing slash)RADARR_API_KEY — API keyRADARR_DEFAULT_QUALITY_PROFILE — optional, defaults to 1The radarr CLI reports a JSON error envelope when they are missing.
Use the installed radarr CLI for common operations. It always emits a single
JSON envelope with ok, command, result or error, and next_actions.
scripts/radarr.sh remains as a compatibility shim for older workflows.
radarr search "Inception" # TMDB lookup
radarr exists 27205 # is it in the library?
radarr add 27205 # add + search
radarr add 27205 --no-search # add but don't search
radarr collection-info 10 # inspect a collection
radarr add-collection 10 # refuses without confirmation
radarr add-collection 10 --confirm-add-collection # whole Star Wars collection
radarr remove 27205 # delete from library, keep files
radarr remove 27205 --delete-files # refuses without confirmation
radarr remove 27205 --delete-files --confirm-delete-files
radarr config # root folders + quality profiles
radarr queue --limit 100 # active downloads
radarr calendar --days 30 # upcoming releases
radarr missing --limit 100 # monitored movies with no file
radarr history --limit 50 # recent history
radarr status # system/status sanity check
For anything not covered, call the API directly with $RADARR_URL and
$RADARR_API_KEY — see references/api-endpoints.md and
references/quick-reference.md.
radarr search "<title>" — present results to user with
[Title (Year)](https://themoviedb.org/movie/<tmdbId>) links. Note any
[Collection: …] tag and offer to add the whole collection.radarr exists <tmdbId> — confirm it's not already there.radarr add <tmdbId> — adds and starts a search.radarr collection-info <collectionTmdbId> — sanity check.radarr add-collection <collectionTmdbId> --confirm-add-collection — adds
everything else and flips monitored/searchOnAdd on so future entries
auto-add.Use the high-level subcommands first (queue, calendar, missing, history,
status). Drop to raw curl for niche queries.
Always confirm with the user before:
remove <tmdbId> --delete-files --confirm-delete-files (irreversible — removes media files)add-collection <collectionTmdbId> --confirm-add-collection on large franchises (can add dozens of movies)/api/v3references/api-endpoints.md — full v3 endpoint reference with
request/response shapesreferences/quick-reference.md — copy-paste curl recipes for common opsreferences/troubleshooting.md — auth, connection, and common error fixes/api/ (v1) paths return 404.X-Api-Key header beats ?apikey= (avoids leaking the key into logs).GET /api/v3/command/{id} polls a command's status.minimumAvailability defaults to released when adding; override in the
payload if you want to grab pre-releases or in-cinema films.RADARR_URL is unreachable, surface that to the
user rather than guessing.development
Generate self-contained HTML visualizations with Plannotator theming. Use for implementation plans, PR explainers, architecture diagrams, data tables, slide decks, and any visual explanation of technical concepts. Plans and PR explainers follow Plannotator's prescriptive approach; all other visual content delegates to nicobailon/visual-explainer.
development
Turn an idea or objective into a goal package for /goal. Interviews the user, builds a reviewed fact sheet via Plannotator, then explores the codebase to produce an execution plan.
development
Open Plannotator's browser-based code review UI for the current worktree or a pull request URL, then act on the feedback that comes back.
testing
Open Plannotator on the latest rendered assistant message and use the returned annotations to revise that message or continue.