Research · Two preprints on Zenodo
On 24 Aug 2026 we filed two preprints on Zenodo. One sets out GRIP, Foxora's written rules for agent work. The other proposes Agentic Language Models, the idea behind Fennec. Both are free to read. Neither is peer reviewed yet.
Filed · 24 Aug 2026 · Zenodo · not peer reviewedTwo papers · two DOIs
Preprint · ZenodoPaper 01
GRIP: The Governed Runtime Intent Protocol
Madhav D. Parashar · Arnav Parashar
Abstract
AI agents act: they call tools, spend money and message people. GRIP makes that work authorised before it runs, bounded at a gate that enforces the limits, and attributable afterwards.
1 · Introduction
DOI10.5281/zenodo.22078457
Preprint · ZenodoPaper 02
Agentic Language Models: A Proposal
Madhav D. Parashar · Arnav Parashar
Abstract
Proposes a class of model whose native interface is the rules of the runtime it serves: a checked package of context goes in, an act the rules can check comes out. And a public test.
1 · Introduction
DOI10.5281/zenodo.22078465
A drawing of the two filed papers, abstracts shortened · a DOI is a permanent address: it always leads to the filed record
01Paper 01 · The protocol
Madhav D. Parashar and Arnav Parashar
AI agents now act. They call tools, spend money and message people. The protocols that connect them say what an agent can reach, but not who allowed the work, how far it may go, or who answers for it. GRIP proposes a small protocol that makes agent work authorised before it runs, bounded at a gate that enforces the limits, and attributable afterwards, whatever the model, framework or vendor.
Inside the paper · six signed objects, one chain
Each object is a small, signed record. Together they answer who allowed the work, how far it could go, and who said done.
02Paper 02 · The model class
Madhav D. Parashar and Arnav Parashar
Most agent systems borrow their judgement from a general-purpose model, then add the rules from outside, as prompts and wrappers. This paper proposes a class of model whose native interface is the rules of the runtime it serves: a checked package of context goes in, an act the rules can check comes out. It also proposes a public test, with pass marks set before any model is measured, so that, once the test code is out, anyone can check whether a model belongs. It is the companion to the GRIP paper.
How we hold our claims
Research is where Foxora is going. These are the rules we keep on the way there.
Both papers are preprints. They propose. They report no results, and no one has peer reviewed them yet.
The model paper fixes its pass marks before any model is measured. Fennec's benchmark results are coming, published with their method so others can check them.
This page may change. Each DOI keeps leading to the same filed record on Zenodo, even if a file moves.
Found a flaw? Write to hello@foxora.ai with the subject “Research”.
Use it today
Foxora Studio keeps the same habits today. Agents ask before going past your brief, stop at your budget, and hand back work you can check. Nothing is done until you accept it.
