Claude Code orchestrates, loomcycle executes — a real 10-agent code review through MCP fan-out (exp7, v0.33.0)

Claude Code orchestrated a 10-agent code review of loomcycle via MCP fan-out, with reviewers running concurrently inside loomcycle and findings consolidated into a report. The review covered 86 files across 10 repo slices, identifying 35 issues including a critical timer leak in internal/channels/scheduler.go that was fixed within hours. The experiment demonstrated the architectural separation where Claude Code serves as the human-facing operator while loomcycle handles multi-agent execution.

Day seven of the operator-via-MCP series — and the cleanest demonstration yet of the architectural shape loomcycle has been driving toward. Claude Code stays the operator and the conversation surface. loomcycle is the side runtime where the actual multi-agent work runs. Topology: a fresh Claude Code session in the jail git-clones loomcycle, then — using its own .claude/agents/code-reviewer.md and .claude/skills/code-review/SKILL.md as the seed — synthesizes a reviewer agent and a code-review skill for loomcycle. One loomcycle import claude-code --from=work/exp7/.claude --write --skills-dest=$PWD/skills later the RFC C2 importer that maps the .claude/ shape onto loomcycle's content-addressed Defs — AgentDef + SkillDef , the operator makes one MCP call: spawn runs N=10, mode=join RFC Y, 464, v0.33.0, shipped today — fanning 10 reviewers across 10 repo slices internal/api/http, internal/tools/builtin, internal/providers, internal/store, internal/config, internal/snapshot, internal/scheduler, internal/pause, internal/channels, cmd/loomcycle . The reviewers run concurrently inside loomcycle, each parking findings in the Memory tool under user scope as a shared ledger. One more spawn run wakes a consolidator that reads the ledger, merges 10 slices into one report, and returns. Result: 10/10 slices, 86 files, 35 issues — 1 Critical + 34 Important. The Critical: internal/channels/scheduler.go:81, a time.AfterFunc closure that can fire before the outer LoadOrStore commits → permanent pendCnt leak under sub-millisecond timer drift. Same-day fixes shipped: 462 + 463 resolved the Critical and most of the Important findings within hours. The Important set surfaced seven structural patterns worth naming: an newID panic on collision no retry , a ToolCtx goroutine leak when the call exceeds context, a restored paused-runs status mismatch, a memory-quota check-then-write race, a MaxBytesReader OOM vector via inflated Content-Length, an interactive-goroutine semaphore leak on early return, and a Refresher.Stop deadlock when the producer holds the same mutex. Three runtime findings surfaced by exp7 itself: Glob abs-path matching falls back to substring on relative roots matches files outside the allowlist ; cross-provider fallback drops reasoning content when the secondary provider doesn't speak the same field; spawn runs with N=10 against a single Anthropic-OAuth subscription tripped the per-key rate limit, surfacing the need for an operator-level fan-out throttle. The substrate-shaped path means the 10 reviewers run as real loomcycle agents, with scheduler reach, memory durability, OTEL spans, and per-run credential isolation, while Claude Code stays the human-facing operator. The contract between the two systems is the MCP wire surface — narrow, structured, well-defined. The substrate becomes the place where multi-agent work actually happens; the operator surface stays human-shaped.