{
 "ledger": {
  "title": "Claim Ledger: Your Multi-Agent System Is an Org Chart",
  "url": "https://durabilitycurve.com/claims/your-multi-agent-system-is-an-org-chart/",
  "essay": {
   "title": "Your Multi-Agent System Is an Org Chart",
   "url": "https://durabilitycurve.com/blog/your-multi-agent-system-is-an-org-chart/",
   "published": "2026-07-07"
  },
  "last_checked": "2026-10-03",
  "format": "current",
  "counts": {
   "claims": 9,
   "by_status": {
    "VERIFIED": 9
   },
   "verified_at_primary": 9,
   "removed": 0
  },
  "license": "https://creativecommons.org/licenses/by/4.0/",
  "attribution": "The Durability Curve (Harry Floyd), https://durabilitycurve.com/claims/your-multi-agent-system-is-an-org-chart/"
 },
 "rows": [
  {
   "id": "MA1",
   "url": "https://durabilitycurve.com/claims/your-multi-agent-system-is-an-org-chart/#c-MA1",
   "status": "VERIFIED",
   "status_label": "Verified",
   "removed": false,
   "claim": "Cognition's June 2025 post, written from its own experience building coding agents and explicitly dated to 2025, said running multiple agents in collaboration produced fragile systems because decision-making ended up too dispersed and context could not be shared thoroughly enough between the agents.",
   "quote": "in 2025, running multiple agents in collaboration only results in fragile systems … decision-making … too dispersed and context isn’t able to be shared thoroughly enough",
   "removal_reason": "",
   "source": "Walden Yan, Cognition, Don't Build Multi-Agents, 12 June 2025",
   "source_url": "https://cognition.com/blog/dont-build-multi-agents",
   "locator": "section \"Applying the Principles\", sub-section \"Multi-Agents\", final paragraph",
   "date_read": "2026-10-03"
  },
  {
   "id": "MA2",
   "url": "https://durabilitycurve.com/claims/your-multi-agent-system-is-an-org-chart/#c-MA2",
   "status": "VERIFIED",
   "status_label": "Verified",
   "removed": false,
   "claim": "On Anthropic's own internal research eval, not an external benchmark, its multi-agent research system (a Claude Opus 4 lead agent with Claude Sonnet 4 subagents) outperformed single-agent Claude Opus 4 by 90.2%.",
   "quote": "Claude Opus 4 as the lead agent and Claude Sonnet 4 subagents outperformed single-agent Claude Opus 4 by 90.2% on our internal research eval",
   "removal_reason": "",
   "source": "Anthropic Engineering, How we built our multi-agent research system, 13 June 2025",
   "source_url": "https://www.anthropic.com/engineering/multi-agent-research-system",
   "locator": "section \"Benefits of a multi-agent system\", second paragraph",
   "date_read": "2026-10-03"
  },
  {
   "id": "MA3",
   "url": "https://durabilitycurve.com/claims/your-multi-agent-system-is-an-org-chart/#c-MA3",
   "status": "VERIFIED",
   "status_label": "Verified",
   "removed": false,
   "claim": "On the BrowseComp browsing evaluation specifically, Anthropic found that token usage by itself explained 80% of the performance variance; three factors together explained 95%.",
   "quote": "three factors explained 95% of the performance variance in the BrowseComp evaluation … token usage by itself explains 80% of the variance",
   "removal_reason": "",
   "source": "Anthropic Engineering, How we built our multi-agent research system, 13 June 2025",
   "source_url": "https://www.anthropic.com/engineering/multi-agent-research-system",
   "locator": "section \"Benefits of a multi-agent system\", third paragraph",
   "date_read": "2026-10-03"
  },
  {
   "id": "MA4",
   "url": "https://durabilitycurve.com/claims/your-multi-agent-system-is-an-org-chart/#c-MA4",
   "status": "VERIFIED",
   "status_label": "Verified",
   "removed": false,
   "claim": "In Anthropic's own data, multi-agent systems used about 15× more tokens than chats, against about 4× for single agents.",
   "quote": "In our data, agents typically use about 4× more tokens than chat interactions, and multi-agent systems use about 15× more tokens than chats",
   "removal_reason": "",
   "source": "Anthropic Engineering, How we built our multi-agent research system, 13 June 2025",
   "source_url": "https://www.anthropic.com/engineering/multi-agent-research-system",
   "locator": "section \"Benefits of a multi-agent system\", fourth paragraph",
   "date_read": "2026-10-03"
  },
  {
   "id": "MA5",
   "url": "https://durabilitycurve.com/claims/your-multi-agent-system-is-an-org-chart/#c-MA5",
   "status": "VERIFIED",
   "status_label": "Verified",
   "removed": false,
   "claim": "Anthropic wrote in June 2025 that domains requiring all agents to share the same context, or involving many dependencies between agents, are not a good fit for multi-agent systems today, and that most coding tasks involve fewer truly parallelisable tasks than research.",
   "quote": "share the same context or involve many dependencies between agents are not a good fit … most coding tasks involve fewer truly parallelizable tasks than research",
   "removal_reason": "",
   "source": "Anthropic Engineering, How we built our multi-agent research system, 13 June 2025",
   "source_url": "https://www.anthropic.com/engineering/multi-agent-research-system",
   "locator": "section \"Benefits of a multi-agent system\", fourth paragraph",
   "date_read": "2026-10-03"
  },
  {
   "id": "MA6",
   "url": "https://durabilitycurve.com/claims/your-multi-agent-system-is-an-org-chart/#c-MA6",
   "status": "VERIFIED",
   "status_label": "Verified",
   "removed": false,
   "claim": "A preprint by Tran and Kiela found that, with reasoning-token budgets held equal, single-agent systems consistently match or outperform multi-agent systems on multi-hop reasoning tasks across three model families (Qwen3, DeepSeek-R1-Distill-Llama and Gemini 2.5). The finding covers multi-hop reasoning only.",
   "quote": "three model families (Qwen3, DeepSeek-R1-Distill-Llama, and Gemini 2.5) … SAS consistently match or outperform MAS on multi-hop reasoning tasks when reasoning tokens are held constant",
   "removal_reason": "",
   "source": "Dat Tran and Douwe Kiela, Single-Agent LLMs Outperform Multi-Agent Systems on Multi-Hop Reasoning Under Equal Thinking Token Budgets, arXiv preprint 2604.02460v2",
   "source_url": "https://arxiv.org/abs/2604.02460",
   "locator": "Abstract (revised 11 April 2026)",
   "date_read": "2026-10-03"
  },
  {
   "id": "MA7",
   "url": "https://durabilitycurve.com/claims/your-multi-agent-system-is-an-org-chart/#c-MA7",
   "status": "VERIFIED",
   "status_label": "Verified",
   "removed": false,
   "claim": "Claude Code's current sub-agents documentation lists its built-in Explore and Plan subagents with read-only tools, Write and Edit denied.",
   "quote": "Explore … Tools: read-only tools; Write and Edit are denied … Plan … Tools: read-only tools; Write and Edit are denied",
   "removal_reason": "",
   "source": "Claude Code Docs, Create custom subagents",
   "source_url": "https://code.claude.com/docs/en/sub-agents",
   "locator": "section \"Built-in subagents\", Explore and Plan tabs, Tools line",
   "date_read": "2026-10-03"
  },
  {
   "id": "MA8",
   "url": "https://durabilitycurve.com/claims/your-multi-agent-system-is-an-org-chart/#c-MA8",
   "status": "VERIFIED",
   "status_label": "Verified",
   "removed": false,
   "claim": "Cognition's April 2026 follow-up said its original observations still hold for parallel-writer swarms, and that the multi-agent setups that work in its practice are those where multiple agents contribute intelligence while writes stay single-threaded.",
   "quote": "Our original observations still hold today for parallel-writer swarms … setups where multiple agents contribute intelligence to a task while writes stay single-threaded",
   "removal_reason": "",
   "source": "Walden Yan, Cognition, Multi-Agents: What's Actually Working, 22 April 2026",
   "source_url": "https://cognition.com/blog/multi-agents-working",
   "locator": "introduction, second paragraph",
   "date_read": "2026-10-03"
  },
  {
   "id": "MA9",
   "url": "https://durabilitycurve.com/claims/your-multi-agent-system-is-an-org-chart/#c-MA9",
   "status": "VERIFIED",
   "status_label": "Verified",
   "removed": false,
   "claim": "Cognition states, as its own company figure, that Devin usage in its largest-enterprises segment, which it describes as traditionally cautious toward new technologies, grew roughly 8× over the six months before its April 2026 post.",
   "quote": "Devin usage in our largest enterprises segment … traditionally been cautious toward adopting new technologies, we see an explosion over the last 6 months (~8x)",
   "removal_reason": "",
   "source": "Walden Yan, Cognition, Multi-Agents: What's Actually Working, 22 April 2026",
   "source_url": "https://cognition.com/blog/multi-agents-working",
   "locator": "section \"What Changed in the Last 10 Months\", first paragraph",
   "date_read": "2026-10-03"
  }
 ]
}