# CLIO

https://akougkas.io/work/clio/

Author: Anthony Kougkas

Updated: 2026-11-12

## What CLIO is

CLIO stands for Context Layer for Input/Output. It is named for the Greek muse of history, because the whole point is that an agent working on science should be able to remember, cite, and account for what it did.

CLIO gives agents what databases give applications: structured, reliable, queryable state. A language model on its own has a prompt and a context window. A scientist has terabytes on a parallel file system, a Slurm queue, instrument metadata, a decade of provenance, and a set of tools with side effects. CLIO is the layer in between.

## How it relates to IOWarp

[IOWarp](https://akougkas.io/work/iowarp/) is the NSF-funded program and platform: the runtime, the engines, and the collaboration among GRC, [The HDF Group](https://www.hdfgroup.org/), and the [University of Utah](https://www.utah.edu/). CLIO is the context architecture I created for it. I set its research direction, define the components and their contracts, coordinate integration across the ecosystem, and answer for what the system promises as a whole.

## What shipped

CLIO is one architecture with technical leadership distributed across the team. I work across its layers while component leads and contributors carry the implementation, research, and releases.

[CLIO Core](https://core.iowarp.ai/) is the tiered context store, with engines that assimilate, transfer, and explore scientific data on the Chimaera runtime. [Luke Logan](https://grc.iit.edu/center/team/luke-logan/) leads its day-to-day engineering. He built the runtime under [Hermes](https://akougkas.io/work/hermes/) as a student, wrote MegaMmap, and finished his PhD in 2025.

[CLIO Agent](https://clio.iowarp.ai/) is the science agent, a desktop and web workspace where a scientist works with data, tools, and papers alongside it. [CLIO Relay](https://github.com/iowarp/clio-relay) lets a desktop tool submit work to a remote cluster and bring back logs, artifacts, and provenance, and the [GACT TUI](https://github.com/iowarp/gact-tui) drives any compliant agent backend from a terminal. [Jaime Cernuda](https://grc.iit.edu/center/team/jaime-cernuda/) leads all three. He built HStream and Hades, defended in 2025, and thinks in pipelines.

[CLIO Kit](https://toolkit.iowarp.ai/docs/intro) collects the scientific tools and Model Context Protocol servers that agents call, from [HDF5](https://toolkit.iowarp.ai/docs/mcps/hdf5) and [Slurm](https://toolkit.iowarp.ai/docs/mcps/slurm) to ParaView and the [National Data Platform](https://iowarp.ai/applications/national-data-platform/). The students [Isa Muradli](https://grc.iit.edu/center/team/isa-muradli/) and [Shazzadul Islam](https://grc.iit.edu/center/team/shazzadul-islam/) build it, and Jaime and I contribute across its interfaces.

[CLIO Coder](https://akougkas.io/work/clio-coder/) is the component I build and maintain myself: a coding agent for the software science runs on, documented at [coder.iowarp.ai](https://coder.iowarp.ai/), and my hands-on testbed for typed tools, bounded delegation, local models, and provenance.

[CLIO Author](https://github.com/iowarp/clio-author), a team-built multi-agent harness for the scientific-paper lifecycle, [CLIO Web Search](https://github.com/iowarp/clio-web-search), and the [Agent Marketplace](https://github.com/iowarp/clio-agent-marketplace) extend the same public ecosystem.

## The loop that is coming

Two more components close the circle. [CLIO Researcher](https://github.com/iowarp/clio-researcher) is an unattended research loop for systems code and configuration: a person writes the directive, a planner model proposes one falsifiable change, CLIO Coder makes it in an isolated checkout, and a grader written as code decides whether the gain holds on a holdout split the planner never sees. A model never grades, and only a person merges the result. CLIO Bench is the record every run lands in, from Researcher and from the evaluations of every other component.

The first campaigns, in September 2026, tuned HDF5 chunk sizes and promoted one change each. Researcher is in public alpha, and Bench is unreleased.

## What carried forward

CLIO inherits its working parts from the systems before it. [The transfer engine](https://core.iowarp.ai/technologies/context-transfer-engine.html) in Core descends from Hermes. The task pipeline was reimplemented from [DTIO](https://akougkas.io/work/dtio/) in 2025. [ChronoLog](https://akougkas.io/work/chronolog/) brought [its MCP server](https://toolkit.iowarp.ai/docs/mcps/chronolog) in-tree with version 3.0 in May 2026, so a CLIO agent can record and replay time-ordered activity through Kit.

## CLIO in the Genesis Mission

In July 2026 the Department of Energy [selected](https://science.osti.gov/-/media/funding/pdf/Awards-Lists/2026/GM-RFA-Awards-List.pdf) two first-round Genesis Mission projects that build on CLIO. [PORTUS](https://grc.iit.edu/research/projects/portus/), led by [Nathan Tallent](https://www.pnnl.gov/people/nathan-tallent) at [PNNL](https://www.pnnl.gov/), builds workflow performance models that guide coding agents as they port and tune scientific workflows on the Genesis Mission Platform; CLIO supplies [the context and data services those agents use](https://iowarp.ai/applications/genesis-portus/). [SPOTTER-AI](https://grc.iit.edu/research/projects/spotter-ai/), led by [Bogdan Nicolae](https://www.bnicolae.net/) at [Argonne](https://www.anl.gov/) with [Oak Ridge](https://www.ornl.gov/) and HPE, captures provenance across Genesis workflows to trace and attribute threats to data and results; GRC [contributes CLIO Core, Kit, Agent, and Coder](https://iowarp.ai/applications/genesis-spotter/). Both are Phase I projects that began in August 2026.

## The bet

The bet is that context deserves systems infrastructure of its own, and that the interesting engineering is in the contracts between components rather than inside any one agent loop. What can be discovered. What can be transferred without copying. What is durable across sessions and machines. What requires a person to approve. What leaves a receipt.

Every component answers those questions differently and ships on its own rhythm. The shared contracts are what keep them one architecture.

## What I am pushing on now

- Making the receipt a first-class object across components, not only inside CLIO Coder.
- Keeping local models the design target everywhere, so a lab with one GPU is a first-class user.
- Getting the shared record schemas (clio-schemas) to carry the contracts I keep describing in prose.
- Building a shared regression corpus so “the agent got better” becomes a measurable claim.
