rule-agent-concurrencylisted

# Agent concurrency Claude Code's own ceilings are far higher than what is useful here: subagents default to **20 concurrent** (`CLAUDE_CODE_MAX_CONCURRENT_SUBAGENTS`), nesting runs **3 deep** (`CLAUDE_CODE_MAX_SUBAGENT_SPAWN_DEPTH`), the per-session spawn cap was removed entirely in 2.1.224, and a workflow runs `min(16, CPUs − 2)` agents at once. Nothing stops a fan-out from exhausting a usage window in one turn. These are the project's own limits. They are lower on purpose. ## Caps | Mode | Ceiling | Use when | |---|---|---| | **Ultra / deep work** | **5–6 agents** | An audit, a multi-lens review, a wide research sweep. This is the maximum for any single fan-out. | | **Normal** | **3–4 concurrent** | Everything else, including background sessions running in parallel. | Count *concurrent*, not total. Ten agents run three at a time is normal mode; six at once is ultra. If a plan needs more than six, it needs a second wave, not a bigger wave — and a second wave means the first one's results narrow the second, which is usually a better result anyway. **Never nest fan-outs.** A depth of 3 means six agents each spawning six is thirty-six live agents. Set `CLAUDE_CODE_MAX_SUBAGENT_SPAWN_DEPTH=1` if a task tempts you. ## Model and effort Pick one of these three. Do not invent combinations. | Tier | Use for | |---|---| | **Opus 5 · xhigh** | The hardest single agent in a fan-out — the judge, the synthesiser, the adversarial verifier. | | **Opus 5 · high** | The default for