FAVUR

2 penalties, 1 spend spike and 1 scale readout

auto-assembled · drafted mechanically from this run's detected moments

Auto-assembled from the 4 sharpest detected moments of this run. 2 of 4 turned; the rest never did.

Or grab the whole reel as one image: composite.png · or as a looping animation: reel.webp

Animated reel: 2 penalties, 1 spend spike and 1 scale readout. Auto-assembled from the 4 sharpest detected moments of this run. 2 of 4 turned; the rest never did. 4 moment cards with source code and transcript excerpts from the run telemetry.
The animated reel - the cards plus source code and transcript excerpts from the run telemetry. The full-size cards follow below.
  1. At peak, 2 agents were mid-tool-call in the same second. One run, one job - and a headcount no single-agent tool can log. 94 agents were hired across 1 hours. A dedicated supervision lane spent 133 model calls doing nothing but checkups on the rest of the team.

    1/4 94 agents hired for one job - a headcount no single-agent tool can log.

  2. Docked 10 merits for a delegation missing tool access - earned back 5 by fixing it. Favur's orchestrator runs periodic checkups on every agent on the team, awarding merits for good work and demerits for bad habits. A checkup caught the develop agent handing work to a child agent without the tool access the job needed, costing it 10 merits - about a fifth of its standing. The agent corrected course and earned 5 merits back before the run ended.

    2/4 The develop agent is docked about a fifth of its standing for a delegation missing tool access. It climbed back.

  3. Docked 10 merits for a delegation missing tool access - earned back 5 by fixing it. Favur's orchestrator runs periodic checkups on every agent on the team, awarding merits for good work and demerits for bad habits. A checkup caught the sprint-review agent handing work to a child agent without the tool access the job needed, costing it 10 merits - about a fifth of its standing. The agent corrected course and earned 5 merits back before the run ended.

    3/4 The sprint-review agent is docked about a fifth of its standing for a delegation missing tool access. It climbed back.

  4. The run's priciest five minutes - 2.3 times a typical window. Every model call is metered, so a run's spend can be read window by window against its own baseline. For five minutes, 71 model requests from 7 agents landed at once - 2.3 times the run's typical five-minute spend. The burst bought something: the checkup agent finished its job.

    4/4 The priciest five minutes runs 2.3 times the typical window - 7 agents burning at once.