Strategic Intelligence

Agent Swarm

The full 11-agent strategic team. Every important assignment gets 10 parallel thinkers, a cross-read pass, and an independent maker-checker before anything ships.

9
Runs completed
78%
PASS rate
11
Agents on roster
β€”
Cost tracking
not instrumented
Live standings
Agent Leaderboard β†’
Storyteller leads (235 pts Β· T2). Run 005: Debrief, Architect, Listener all cross into T2.
πŸ†

Run Archive

#010Vercel Deep DivePASS

The system was further along than expected. The `/portal/[client]` page already had summary cards and a stage timeline. A separate page at `/deals/hr-com/meetings` already had 268

architectaudit-qualityhunterstorytellerdebrief (Phase 1 parallel research) + conductor synthesis for build phases
2026-05-04
Phase 1 research delegated to 5 specialized subagents; conductor played remaining agent roles synchronously due to session-bound execution model
View archive β†’
#009Agent Swarm Mode Feature SystemPASS

The Agent Swarm Mode feature system spans 107+ files across the monorepo: 53 files in the maxswarm skill alone, 20 in swarm-upgrade, 2 Next.js pages, 7 public HTML archives, 14 eng

data-architectstorytellerarchitect (system-map)audit-qualitytech-translator
2026-05-04
5-agent subset: pure internal audit + classification β€” no buyer research, no client-facing copy, no deal context needed
View archive β†’
#008Salesfinity Weekly PipelinePASS

There are no calls to analyze. Across all four data sources β€” Salesfinity, Fireflies, SpokePhone, and Gmail β€” zero call records were found for the period 2026-04-27 to 2026-05-04.

listenerquarterbackwriterdebriefaudit-qualitytech-translator
2026-05-04
Pipeline ops assignment β€” no contact discovery (hunter), no emotional copy (draper), no market sizing (market-analyst), no narrative coherence needed (storyteller), no infra implications (architect).
View archive β†’
#007Repo Consolidation One RepoPASS

Consolidating saves ~$12k/year for a 1-engineer team, with a six-week payback. The plan works, but three high-risk steps (the Lob webhook re-pointing, the deals Supabase migration,

architectquarterbacklistenermarket-analyststorytellerwriter (Phase 1)+1 more
2026-05-03
`pure-strategy` preset β€” internal infrastructure decision; no contact discovery (hunter excluded), no client emotional layer (draper excluded). Listener covered intern unstated needs in lieu of draper.
View archive β†’
#006Hrcom Data Room BlueprintPASS

The swarm produced a complete 38-document data room organized into 8 sections, with every document containing a summary statement, a "Why This Matters" explanation, numbered fill-i

architectwritermarket-analystdraperhunterlistener+2 more
2026-05-01
Data room construction β€” storyteller, tech-translator, debrief excluded (writer handled plain English + narrative; debrief Phase 6 only)
View archive β†’
#005Listener Migration Intent AnalysisPASS

2026-04-30
full
View archive β†’
#004Enrichment Audit Post Run003PENDING

One signal was fully deployed. Three were not. The one that was fixed (caching confirmed sponsors in Supabase) was fixed today β€” in this session β€” only because Ewing directly asked

storyteller (co-lead)quarterback (co-lead)+ 3 voted: architectaudit-qualitydebrief
2026-04-30
Targeted audit panel β€” storyteller/quarterback co-leads per Ewing directive; voted in architect (system state), audit-quality (independent verdict), debrief (learning extraction). hunter/writer/draper/market-analyst/listener/tech-translator excluded β€” no contact discovery, no copy, no market sizing needed.
View archive β†’
#003Deal Tracker Deal RoomPASS

The deal-tracking system was eventually shipped in `next-chapter-os/src/app/room/[slug]/` as the deal-room route, with `v_deal_room_content` view and `deal_plays` table populated f

2026-04-30
full
View archive β†’
#002Hrcom Buyers Exa MasteryPENDING

(filled in Phase 6)

full 11 (writerarchitecthunterquarterbacklistenerdraper+5 more
2026-04-30
n/a β€” broad strategic + research + code work warrants full slate
View archive β†’

How the swarm works

Phase 1 β€” Parallel drafts

All 10 working agents get the same brief and produce independent answers. No agent sees another's draft yet.

Phase 2 β€” Cross-read & poke holes

Every agent reads all 9 peer drafts, then re-examines their own. They must find where they were wrong before defending their position.

Phase 4 β€” Maker-checker

Audit-quality sees only the synthesized final output β€” never the drafts. Returns PASS / FIX-AND-SHIP / REWORK. Conductor acts on the verdict.

Swarm runs are archived here and in ~/.claude/skills/maxswarm/previous_swarms/ on the Mac Mini.