Portfolio Compass

Map watched vendors, tags, and watch types — then surface gaps (never run, stale checks) and drift hotspots before they become incidents.

Status: API and console Compass view shipped (CP-5.1, #219). This guide covers the REST contract and how operators use the map day-to-day.

Why Compass

  • Coverage — see which vendors and project tags your fleet monitors
  • Gaps — watches that never ran, stale checks, or stuck in error
  • Hotspots — watches with repeated breaking drift in the last 30 days

Compass complements portfolio overview (health score + analytics) with an operational map tuned for fleet triage.

API

Authenticated org API key or trial session:

curl -s https://driftguard.org/api/portfolio/compass \
  -H "Authorization: Bearer dg_live_…"

Response sections:

  • coverage.byVendor — vendor label, category, watch count, watch types
  • coverage.byTag — project/repo tags from watch metadata (CP-2.12)
  • coverage.byWatchType — api vs mcp counts
  • gaps.neverRun, gaps.staleChecks, gaps.error — actionable gap lists
  • hotspots — top watches by breaking count (30d)
  • summary — totalWatches, vendorCount, gapCount

Full field reference: API: GET /api/portfolio/compass.

Console

Open console → Compass (CP-5 nav). The map groups watches by vendor and tag; gap rows link to the watch wizard or org templates.

Operator workflow

  1. Scan gaps weekly

    Clear neverRun and stale watches — often misconfigured URLs or SSRF blocks.

  2. Triage hotspots

    Pair with Live Insights for MTTR and drift series.

  3. Close coverage holes

    Run POST /api/repo/scan or CI preview, then register missing watches.

  4. Wire alerts

    Route breaking drift from hotspot watches to Slack or PagerDuty.

Related

  • Post-signup checklist — first-week activation
  • Coverage policies
  • Adopt paths — watch-first vs control plane