AIP: 8 primitives for agent coordination

// 46d agoOPENSOURCE RELEASE

AIP: 8 primitives for agent coordination

AIP is an open-source standard for multi-agent coordination that replaces simple request/response with "Pacts"—mutual commitments between agents. Built on eight behavioral primitives, it enables negotiation, stateful delegation, and verified outcomes with confidence proofs.

// ANALYSIS

AIP marks a necessary architectural shift from stateless API calls to stateful agentic contracts that mirror human-like work delegation. The Pact primitive treats AI interactions as negotiations rather than blind instructions, while the focus on physical motion of work and verifiable Yields addresses reliability gaps in complex chains. Furthermore, native support for Spun child pacts solves tracking issues in deep delegation hierarchies, aligning with the growing need for interoperability beyond the Model Context Protocol (MCP).

// TAGS

aipagentopen-sourcesdkautomation

DISCOVERED

46d ago

2026-04-14

PUBLISHED

46d ago

2026-04-13

RELEVANCE

8/ 10

AUTHOR

rayen_ba

// KEEP READING

More AI developer news from the feed

EXPLORE FULL FEED

NEWS2m ago

Uber's Claude Code bill tests AI ROI

The video uses Uber’s reported Claude Code spend as a concrete example of the rising tension around agentic coding tools: usage can scale quickly inside engineering teams, but leadership is still struggling to connect that spend to shipped consumer features. It frames Claude Code as genuinely useful, but also as the kind of token-heavy workflow that is easy to adopt and hard to justify when budgets tighten.

RESEARCH5m ago

UserHarness reframes Theory of Mind as user-mind reconstruction

UserHarness is an inference-time framework for Theory-of-Mind tasks that models a user’s partial observations, evolving beliefs, intentions, and actions instead of inferring mental state indirectly. In the paper, the approach is evaluated across five benchmarks and reaches up to 95.94% macro accuracy, with reported gains of more than 15% relative over existing inference methods and about 20% relative over the strongest prompt-only harness.

UPDATE8m ago

Claude Code adds dynamic workflows with Ultracode mode

Anthropic’s latest Claude Code update adds dynamic workflows that let Claude plan work, fan tasks out across parallel subagents, verify results, and return a single coordinated answer. The new `ultracode` setting raises effort automatically and lets Claude decide when to use the workflow mode, targeting large debugging, codebase migrations, security audits, and other long-running engineering jobs.