OpenIntent Protocol proposes declarative standard for AI agent interoperability

·1 views

The OpenIntent Protocol (OIP) is a declarative communication protocol designed to improve interoperability between AI agents built by different organizations, languages, and infrastructures. Currently in development as part of the H2A2H architecture (v0.3), the protocol is still evolving ahead of a planned v1.0 release. OIP's core principle is that an agent should declare what it wants to accomplish rather than how to communicate, with the infrastructure handling transport details. The protocol is structured around four components — intent, capability, channel contract, and interaction patterns — supporting models such as request-response, streaming, event-driven, broadcast, and fire-and-forget. By decoupling application logic from infrastructure, OIP aims to eliminate the need for agents to be modified when underlying communication protocols such as gRPC, NATS, or QUIC change.

Read the full story at DEV Community

This is an AI-generated summary. ShortSingh links to the original source for the complete article.

Discussion (0)

Developer's deploy scripts passed all checks while shipping broken code for weeks

A software developer discovered that their deployment pipeline consistently reported success while silently failing to apply critical code changes in production. A Celery Beat scheduler container was excluded from the deploy script, meaning it ran two-week-old code through multiple deployments without any alert. A separate incident saw an annual billing feature ship dead because new environment variables were added after containers had already started. All automated checks passed in both cases because they verified general server state and uptime, not whether the intended changes had actually taken effect. The developer notes the core problem is a gap between checks that confirm system health and checks that confirm deployment intent.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

Open-Source Runtime Bartholomew Brings Sub-Microsecond Safety Guards to AI Agents

Developers building autonomous AI agents with frameworks like LangChain, AutoGen, and CrewAI often face risks when granting large language models direct control over filesystems, databases, and OS interfaces. To address this, a team has launched Bartholomew (BTP v2.5), an open-source execution runtime and MCP proxy for Python and Node.js. The tool applies four core safety primitives: sub-microsecond OS event gating at 0.95 µs, in-memory Copy-on-Write rollbacks that restore filesystem state within milliseconds if a tool call violates boundaries, recursive token quota attenuation to prevent runaway multi-agent API spend, and bi-directional secret scrubbing at 0.82 µs to block credential leakage in both directions. Unlike existing approaches such as remote moderation filters or container virtualization, Bartholomew treats agent execution as an atomic, reversible micro-transaction inspired by database transactional theory. The project is available as open source and supports common credential patterns including OpenAI, Anthropic, AWS, and GitHub tokens.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

How one team built 400 lines of code to safely let AI query customer databases

A software team built an AI agent that reads support tickets, queries customer databases, and returns findings with SQL evidence attached. Rather than relying on prompt instructions to restrict the AI's behavior, the team enforced every safety rule through code that parses and inspects the generated SQL as an abstract syntax tree (AST). The system rejects anything other than a single, plain SELECT statement and blocks a denylist of dangerous SQL node types — including constructs like SELECT INTO and FOR UPDATE that technically parse as SELECT but can write data or acquire locks. All table references are validated against an allowlist by traversing the full AST, catching hidden reads inside CTEs, subqueries, and UNION clauses that naive string checks would miss. The codebase is open-source under the Apache 2.0 license, and the team notes the approach was refined after getting it wrong twice before arriving at the current design.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

How Shapezo Fits Into Game Engine Pipelines as a Level Design Intent Tool

Shapezo is positioned as a pre-production spatial composition tool that helps game development teams define level design intent, landmark hierarchy, and gameplay routes before engine constraints take over. It is not meant to replace game engines, DCC tools, or source control, but instead acts as a structured hypothesis layer covering terrain, architecture, and circulation. Teams are advised to assign each concern to the appropriate system — Shapezo for context and layout, the engine for runtime physics and navigation, and DCC tools for optimized meshes and materials. Large environments benefit from using Shapezo to define districts, sightlines, and streaming boundaries before individual assets are authored. Playtest findings should feed back into the spatial model, allowing teams to revise overall composition rather than expand asset scope prematurely.

0 comments Read more at DEV Community