Hosted · /team/brain-metrics

Team brain metrics

Your team's brain, summed. Errors-per-week across every tracked repo. Recurring prevention-rule fingerprints (a rule that failed twice is a rule that needs work). Decisions-per-month pace. Aged open questions. The institutional-memory health view, rolled up from every bound desktop.

See it in motion

Where to find it

  • Hosted: https://repoops.ai/team/brain-metrics
  • Source: Each bound desktop streams a per-repo brainPulse() snapshot hourly via the brain-pulse streamer.
  • API: POST /api/brain-pulse/ingest (desktop → hosted, device-token auth).
  • Localhost equivalent: The Today cockpit's Brain pulse panel + the Brain reflection tab.

What it does for you

You'll see the brain compound.Errors-per-week trending down. Decisions logged climbing. The team memory isn't a slogan — it's a number that moves.
You'll catch the rules that failed.Recurring prevention-rule fingerprints surface here. A rule that triggered twice in 30 days means the rule isn't doing its job — surface it, fix it, or replace it.
You'll see aged open questions.Questions filed > 30 days ago with no resolution show up as aged. Either resolve them, escalate, or accept they're permanently parked.

Configure

Nothing on the hosted side. The desktop streamer needs the BYOK Anthropic key (for the reflector) and the bound-device token; both are set during the linking-code flow.

  • REPOOPS_BRAIN_DREAM (desktop) — kill switch on the reflector itself. The streamer ships regardless (it streams metric summaries, not LLM output).
  • Capture mode (server-owned policy): brain-pulse always streams in metadata mode — no proposal bodies are sent. Fingerprints + counts only.
  • Streamer cadence: hourly while online. The desktop's supervise.mjs single-process timer.

Use it well

  1. Look at errors-per-week trend first.

    If it's climbing, the team's prevention rules aren't firing. Open the recurring-rule fingerprint list — that's where the leverage is.

  2. Pick one recurring-rule fingerprint to fix per week.

    The rule already exists in CLAUDE.md; the issue is that it failed. Either re-word it, add an example, or replace it with a checkable rule.

  3. Close aged open questions weekly.

    30+ days = either resolve or move to backlog. Aged questions are a quality signal, not a TODO list.

  4. Decisions-per-month is a leading indicator.

    Climbing = team is making architecture calls and documenting them. Flat = team is moving by default. Either is fine; just be honest about which.

Read more