Menu
One-glance health overview of every media service (radarr, sonarr, sportarr, prowlarr, bazarr, qbittorrent, nzbget). Hits each service's status/health endpoint and reports version, reachability, and any active health warnings.
Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Basé sur la classification professionnelle SOC
| name | media-status |
| description | One-glance health overview of every media service (radarr, sonarr, sportarr, prowlarr, bazarr, qbittorrent, nzbget). Hits each service's status/health endpoint and reports version, reachability, and any active health warnings. |
| version | 0.1.0 |
| author | dobry-ops |
| license | MIT |
| metadata | {"hermes":{"tags":["media","status","health","diagnostics"],"category":"media-ordering","related_skills":["arr-search","qbit-list","nzbget-list"]}} |
Read-only health check across the whole media stack. Use this first whenever the user reports "something's broken" with media — it tells you which service is the prime suspect.
All of:
$RADARR_URL, $RADARR_API_KEY,
$SONARR_URL, $SONARR_API_KEY,
$SPORTARR_URL, $SPORTARR_API_KEY,
$PROWLARR_URL, $PROWLARR_API_KEY,
$BAZARR_URL, $BAZARR_API_KEY,
$QBITTORRENT_URL, $QBITTORRENT_USERNAME, $QBITTORRENT_PASSWORD,
$NZBGET_URL, $NZBGET_USERNAME, $NZBGET_PASSWORD.
For each *arr app — version + health:
for app in radarr sonarr sportarr prowlarr bazarr; do
URL_VAR="$(echo $app | tr a-z A-Z)_URL"
KEY_VAR="$(echo $app | tr a-z A-Z)_API_KEY"
URL="${!URL_VAR}"; KEY="${!KEY_VAR}"
echo "=== $app ==="
curl -fsS --max-time 3 -H "X-Api-Key: $KEY" \
"$URL/api/v3/system/status" \
| jq '{version, branch, runtimeVersion, isProduction}' 2>&1 \
|| echo " ✗ unreachable"
curl -fsS --max-time 3 -H "X-Api-Key: $KEY" \
"$URL/api/v3/health" \
| jq 'map({source, type, message})' 2>&1 \
|| echo " ✗ health endpoint unreachable"
done
qBittorrent — version + torrent counts by state:
COOKIE=$(mktemp); trap 'rm -f "$COOKIE"' EXIT
curl -sf -c "$COOKIE" --max-time 3 \
-d "username=$QBITTORRENT_USERNAME&password=$QBITTORRENT_PASSWORD" \
"$QBITTORRENT_URL/api/v2/auth/login" > /dev/null
curl -sf -b "$COOKIE" --max-time 3 "$QBITTORRENT_URL/api/v2/app/version"
curl -sf -b "$COOKIE" --max-time 3 "$QBITTORRENT_URL/api/v2/torrents/info" \
| jq 'group_by(.state) | map({state: .[0].state, count: length})'
NZBGet — status method:
curl -sf --max-time 3 -u "$NZBGET_USERNAME:$NZBGET_PASSWORD" \
-H 'Content-Type: application/json' \
-d '{"method":"status","params":[]}' \
"$NZBGET_URL/jsonrpc" \
| jq '.result | {ServerPaused, DownloadPaused, RemainingSizeMB, DownloadRate, FreeDiskSpaceMB, ServerStandBy}'
=== Media stack ===
radarr ✓ v6.1.1 no health issues
sonarr ✓ v4.0.15.2 1 warning: indexer "X" is unreachable
sportarr ✓ v4.0.15.2 no health issues
prowlarr ✓ v1.42.0 no health issues
bazarr ✗ unreachable (timeout)
qbittorrent ✓ v5.0.5 34 downloading, 412 seeding
nzbget ✓ 2 active, 412 MB queued, 28 GiB free disk
Lead with the unhealthy ones if any. If everything's green, say so plainly in one line.
--max-time 3 on every request so a single dead service doesn't hang
the whole report.Poll trusted Multiple Sclerosis RSS feeds, score credibility, save raw items, filter low-quality material, and promote high-value items into citation-grounded KB processing.
Save a URL, file path, pasted text, DOI, or PMID into the knowledgebase at $KB_ROOT/raw/manual/YYYY/MM/DD/. Enriches via the relevant MCP (pubmed/crossref) when input is a DOI or PMID. NON-DESTRUCTIVE — write-only into raw/, never into pages/.
Append a single timestamped bullet to today's $KB_ROOT/journals/YYYY/MM/YYYY_MM_DD.md describing what the agent just did. Logseq-compatible and date-partitioned. Call this at the end of every meaningful action (page write, ingest, query answered).
Self-heal the MS knowledgebase layout after RSS watch or weekly digest. Migrates legacy flat files into the V2 date/type structure, refreshes Start Here and Index pages, checks links/frontmatter, and journals the maintenance summary.
Answer a research question by searching the kb first, then PubMed → CrossRef → SearXNG, writing a citation-grounded content page under $KB_ROOT/content/<type>/YYYY/MM/. Every claim cites a DOI/PMID/URL or is dropped. This is the heart of the agent.
Safely prune generated RSS watcher raw cache after weekly digest retention, without touching manual raw ingests or the RSS seen ledger.