The main agent just hired its own team. anthropic's dynamic workflows for managed agents, in public…
the main agent just hired its own team. anthropic's dynamic workflows for managed agents, in public beta since oct 9, let one agent assign sub-agents — and on 116,000 lines of code the swarm found 66 bugs where a single agent found 27.
the handoff now happens inside the model.
Context
Verified event date: Anthropic moved dynamic workflows for Claude Managed Agents to public beta on Oct 9, 2026 (The Decoder, CellCog and Chasing Next, Oct 9). A lead agent writes a plan, splits the job among subagents in phases and merges the results; a run can use up to 1,000 agents over its life, with up to 64 working at once.
In Anthropic's own test, a codebase of 116,000 lines had 70 planted bugs. A single agent found 14, 15 and 27 across three runs, while the dynamic workflow found 66 in each run (94.3 percent recall).
The Oct 9 public beta, the 116,000 lines, 66 bugs and the single-agent best of 27 match the reports. The post leaves out that the test had 70 planted bugs and that single-agent runs found 14, 15 and 27, so 27 is the single agent's best run, not its typical result. The test is Anthropic's own. 'The handoff now happens inside the model' is the author's; the sources describe a workflow program that Claude writes and Anthropic's server runs.
'the main agent just hired its own team' is the author's opinion.
Related work
Watch next
- Check how the 70-bug test was planted and scored in Anthropic's own write-up.
Sources
- Claude Docs, Workflow runsplatform.claude.com
- CellCog, Claude Dynamic Workflows, Oct 9, 2026cellcog.ai
- AlphaSignal, dynamic workflowsalphasignal.ai
Provenance
The note above is reproduced unedited from the original post, first published on Threads on 11 October 2026 at 11:35 IST. Sources are the papers and datasets the note draws on.
View the original post ↗Embed this note
More notes
The air is now being asked to keep its own ledger
the air is now being asked to keep its own ledger: ecmwf’s aifs compo becomes the first ai model to forecast atmospheric composition globally every three hours, cleanair simulates 365 days of pm2.5 over china in ten seconds, and a unified framework maps six pollutants at one kilometer across the whole country. the air now files its own composition report.
read the note →The current is now being asked to draw its own map
the current is now being asked to draw its own map: china’s langya 2.0 predicts six ocean phenomena including internal waves and mesoscale eddies, a deep net called wenhai resolves eddies globally with air sea flux formulas built in, and scripps infers surface currents from the way temperature patterns deform in satellite images. the ocean now files its own circulation report.
read the note →The soil is now being asked to report its own carbon
the soil is now being asked to report its own carbon: a nix color sensor paired with generative data augmentation predicts soil organic carbon without a lab, random forest drives 74 percent of soil health mapping studies, and sentinel 2 tracks five year carbon change across france and italy from 922 samples. the dirt now files its own carbon account.
read the note →