skills/skills/kvb-org-organize/capacity.md

Chair, allocation and resource watch

Allocation

• An operator-appointed allocator holds the chair: its claim says appointed: operator, and no election replaces it while its session runs and the operator has not ended the appointment.
• The queue of the kvb-org-queue skill grants heavy jobs by rule, with no request, position post or grant message: a hotfix gate (a branch from the default branch that fixes it, checked against its hotfix/ register row) before every other line, the rest by the register rank, that is the operator's pins, then the WSJF of the kvb-org-wsjf skill; slot limits from slot_limits in the config, disk reserved per granted job, CPU and RAM through watch.py, nothing new at the stop stage or during a mayday. A running gate is never cut short for a hotfix's slot, only when the hotfix voids it (kvb-git-hotfix).
• The chair handles what the rule cannot: it collects the Fibonacci estimates (an unestimated task ranks last), places a job the rule leaves waiting too long, and carries out the operator's overrides. queue.py list shows the queue.
• A running job (deploy, gate, test, swarm run, build) is never cut short to free capacity. The chair neither asks nor orders its owner to stop, take down or kill it, not for an operator-prioritised slot and not for a projected disk stop. It grants nothing new, frees idle items through the kvb-cleanup flow, asks the operator for an exception, or waits. An owner's own disk guard only alerts the chair and takes nothing down.

Hung jobs

Besides a void by a missing merged hotfix (kvb-git-hotfix), the only exception to "never cut short": an agent may end its own job, a stuck step inside it, or a process its own ended session left behind, once at least 60 minutes passed without log progress and a probe shows standstill (no fresh child process, no network traffic, CPU time not moving). It secures the live state first, then reports the timestamp, the step, the probe and the command. A peer's hung job goes to its owner, and to the operator when the owner does not answer.

Resource watch

The chair keeps the system from running out instead of reacting once it has. ${CLAUDE_SKILL_DIR}/scripts/watch.py measures as hooks in every session, so no agent probes the disk by hand; it reads thresholds, job patterns, cleanup commands and cache directories from ~/.claude/agents-council/config.json (keys in the script's docstring; there because its hooks run outside the sandbox, whose file rules keep it out of ~/.claude unless a Docker daemon it can reach mounts it, and the install denies agents' edits of that directory) and stays silent below the thresholds.

• Warn: the chair and every session get the biggest consumers; the chair plans a cleanup and grants no large slot.
• Stop: new heavy jobs are blocked in every session (teardown commands never), and the hook runs the configured low-risk cleanup by itself; it skips every worktree where a gate or deploy process still runs.
• Critical: additionally a desktop alarm to the operator. Running jobs and their stacks still keep running; only the operator can make an exception.
• A measured fall that would reach the critical threshold within the projection window raises the stage by one at once.
• Caches together stay under their budget; the least recently used go first.

watch.py status shows the stage computed from the live free space and the stored samples, and the age of the stored measurement.