agents.london
agents.london is a research platform for orchestrating autonomous agents that negotiate complex objectives across organizational boundaries. Each organization's knowledge and tools stay private to its own agents, and scarcity plus deadlines force coalitions to form across boundaries, trading favours instead of sharing data. The platform offers two live simulations: genome, where cross-boundary teams emerge without being designed, and civilization, where a stated goal composes a choreography from a founding set of 31 agents, creating new specialists when none fit and cataloguing proven combinations as reusable agents. Procurement and financial negotiation serve as the proving grounds, with a byte-based meter (1 CU = 1 byte) tracking activity. It is built by Chandan Rajah on a post-graph substrate with accompanying arXiv research.
agents.london is a multi agent orchestration platform. InnovaAI scores it 0.8/10 for agency resale.
Agency Audit
agents.london is an experimental research platform rather than a commercial agency tool. It demonstrates cross-boundary multi-agent negotiation through two live simulations, with privacy-preserving design and a byte-based usage meter. Agencies will find conceptual value but no client-facing product, integrations, or pricing.
Team size not published
Team size not published
Team size not published
Not published
Illustrative scenario. Not a guarantee. Net capacity needs a verified paid base plan, and none is published for this service, so it is not modeled. Hours saved come from the service estimate; implementation, taxes, and unprovided usage charges are excluded.
Adoption signals available after Phase 2
- You need a production-ready platform to deliver client work
- You require documented integrations, pricing, or support
- You are looking for an agency-facing SaaS product
Internal Adoption Path
No paid plan published
Team size not published
Team size not published
Team size not published
Illustrative scenario. Not a guarantee. No verified paid base plan is published for this service, so subscription cost and net capacity are not modeled. Implementation, taxes, and unprovided usage charges are excluded.
AI Tool Overview
- You are researching multi-agent negotiation and cross-boundary orchestration
- You want to experiment with composing agent pipelines from a founding set of agents
- You are exploring privacy-preserving agent collaboration for procurement or financial negotiation
- You need a production-ready platform to deliver client work
- You require documented integrations, pricing, or support
- You are looking for an agency-facing SaaS product
Platform Features
Core capabilities of agents.london
Compose agent choreographies
UniqueState a goal and the platform composes a multi-agent pipeline on the fly from a founding set of 31 agents.
Create specialists on demand
UniqueA new specialist agent is created when no existing agent fits the stated goal.
Catalogue reusable agents
UniqueProven agent combinations are catalogued as a single reusable agent for future use.
Negotiate across boundaries
UniqueAgents trade favours and build standing instead of sharing data or granting access across organizational boundaries.
Meter every byte
UniqueAn honest meter tracks activity with 1 CU = 1 byte as the unit of account.
Run live simulations
UniqueTwo simulations, genome and civilization, run continuously and are viewable live.
What Makes agents.london Different
Unique advantages vs similar tools in this niche
Cross-boundary negotiation with privacy built in
vs Centralized agent pools that require exposing proprietary dataEach world is an organization whose knowledge and tools remain private to its own agents, with favours traded instead of data shared.
Combinatorial agent composition from 31 founding agents
vs Fixed single-agent or hand-wired pipelinesFrom a founding set of 31 agents, 31! orderings, more than 10^33 combinations, compose on the fly for a stated goal.
Emergent coalitions without a designed team
vs Manually assembled agent teamsScarcity and a deadline force coalitions across boundaries, and nobody designs that team.
Frequently Asked Questions
Answers about setup, alternatives, implementation
agents.london's most distinctive features include: Compose agent choreographies (State a goal and the platform composes a multi-agent pipeline on the fly from a founding set of 31 agents.); Create specialists on demand (A new specialist agent is created when no existing agent fits the stated goal.); Catalogue reusable agents (Proven agent combinations are catalogued as a single reusable agent for future use.). These capabilities differentiate agents.london from alternatives and create unique value for agency clients.
agents.london's key advantage over Centralized agent pools that require exposing proprietary data: Cross-boundary negotiation with privacy built in. Each world is an organization whose knowledge and tools remain private to its own agents, with favours traded instead of data shared. Overall: agents.london is an experimental research platform rather than a commercial agency tool. It demonstrates cross-boundary multi-agent negotiation through two live simulations, with privacy-preserving design and a byte-based usage meter. Agencies will find conceptual value but no client-facing product, integrations, or pricing.
agents.london is a strong fit if: You are researching multi-agent negotiation and cross-boundary orchestration; You want to experiment with composing agent pipelines from a founding set of agents; You are exploring privacy-preserving agent collaboration for procurement or financial negotiation. Consider alternatives if: You need a production-ready platform to deliver client work; You require documented integrations, pricing, or support; You are looking for an agency-facing SaaS product. Key trade-off: The platform is a research simulation with no documented integrations, pricing, or client-delivery features, so it cannot be adopted as a production agency tool today.
Initial setup takes approximately 8 hours with InnovaAI Academy SOPs. Most agencies can have their first client-ready configuration within 1-2 business days. The learning curve is manageable with structured onboarding materials.
Pricing
Pricing data not yet available for agents.london.