open-agent-profile logo

open-agent-profile

Free

An open specification for persisting a named AI agent as a file instead of a running process.

Visit Website
Related by shared tags and keywordsagentic-toolsaxarAn agent harness for local models
ai-toolsagent-frameworkagentic-aiai-agentsinteroperabilityaiai assistantautomation

Overview

Open Agent Profile (OAP) is an open specification for persisting a named AI agent as a file instead of a running process. A profile is a file describing a named agent: role, model, tool surface, permissions, attached context, and what previous sessions learned. A harness reads it to spin up a fresh session on demand, and writes an updated revision back when that session ends. No process is resident. The file is the agent.

The specification defines three safety rules. A profile narrows. It never widens. An agent cannot rewrite its own contract. Learned state is untrusted content. Sessions emit a state delta, and delta operations may only touch /state; changes to tools, permissions, model, or instructions go into a proposals block with a written rationale, and a human approves it.

OAP does not replace Agent Skills, MCP, or harness config. Skills are what an agent knows how to do. MCP is what it can reach. Harness config is what it is allowed to do. OAP is who it is, and what it has learned. Support libraries exist for Python, TypeScript, Go, Rust, and Java, and conformance levels are 1 Read, 2 Read/Write, and 3 Full.

Pricing

Free

Detailed plans have not been confirmed in our catalog. Check the official website for current limits and billing terms.

Visit Website

Prices and limits may change. Confirm the currency, billing period, seat minimum and usage caps on the official website.

Use Cases

  • Teams use OAP to persist a named AI agent as a file instead of a running process.
  • Teams use OAP profiles to define an agent's role, model, tool surface, permissions, attached context, and learned state.
  • Teams use OAP to spin up a fresh session on demand from a profile and write an updated revision back when the session ends.
  • Teams use OAP to apply what a session learned while holding capability requests for human review.
  • Teams use OAP to validate profiles and compute RFC 8785 digests.
  • Teams use OAP to load a profile, assemble the prompt in specification order, and report what it dropped.
  • Teams use OAP to turn a session into a reviewable delta and apply it.
  • Teams use OAP to discover, validate, and instantiate an agent from a profile.

Features

Defines an open agent profile specification

Persists named agents as files

Avoids requiring a running agent process

Makes agent profiles portable

Keeps learned state reviewable

Places bounds on learned state

Reviews

No reviews yet. Be the first to share your experience!

Rankings & collections