Re: interesting post - AI Agent Protocols, the HTTP of AI agents—a shared language for coordination

Hello,

Thanks for the links.

Yes indeed there is already a lot done in industrial applications and also
this group incorporates the work done for the Web Of Things (W3C WOT
https://www.w3.org/WoT/documentation/ ) in the field of IoT connectivity.
Another major role is played by semantic technology like Linked Data (RDF
at large, and Hydra for Web APIs). I have recently joined and am trying to
understand the pieces in play.

In my personal opinion, the main challenge of the group is to provide a
frame of reference for what is happening with all the new NLP applications
that have been built on top of HTTP in the last few years. So while there
are established protocols based on machine-readable formats leveraging
machine-readable ontologies (that is what I understand is developed in the
semantic-specific FIPA documents linked), what is surfacing now is the need
of a *minimal common standard to support a close interaction between
machine-protocols and Natural Language* (mostly made possible by LLMs, see
example below). So, if it is true that the underlying layer can still
leverage all the existing standards, it has now emerged a much heavier use
of the layer that connects human-facing interfaces and automated agents
that is, in my understanding, the focus of this group.
For example: the Model Context Protocol (MCP) is designed by an
industry-player as an open, machine-usable protocol that standardizes how
AI applications (such as LLMs and agents) interact with external tools,
data sources, services or networks. Now, we have quite established
ecosystems for all those targets but we don't have a minimal standard
reference to leverage LLMs to do the management of those resources, as
there are no standard ways for the LLM to communicate with these diverse
resources (that may be using FIPA or any other standard internally). So at
the moment every player is trying to develop its own solution in a very
interesting rush for automation.

An example of potential reuse of FIPA concepts in Web Agents:


> *Bridge FIPA Protocols with MCP/Context Architecture** Mapping FIPA ACL
> to MCP: Define a mapping layer where MCP function calls, resource accesses,
> and prompts are encoded as FIPA ACL messages. For example, an MCP tool
> invocation can be represented as a FIPA ‘request’ performative, with tool
> parameters in the content field.
> * Protocol Extensibility: Allow new performatives or ontologies to be
> registered for LLM-specific actions (e.g., context updates, prompt
> chaining), while maintaining backward compatibility with FIPA’s core set.
> * Coordination Patterns: Reuse FIPA’s interaction protocols as templates
> for orchestrating multi-agent/LLM workflows (e.g., contract net for dynamic
> tool selection, request/inform for context propagation).


It is the objective of the interoperability taskforce, in my understanding,
to collect all the common characteristics of these new tools and all the
shared technologies so to define a minimal framework that every player
willing to define its own of this layer can start from, thus the different
sections in the interoperability document
https://w3c-cg.github.io/webagents/TaskForces/Interoperability/Reports/report-interoperability.html

Please contribute your points about FIPA or other matters at
https://github.com/w3c-cg/webagents/issues

Hope everybody else will barge in with their own understanding.

Regards,

Lorenzo Moriondo
ロレンツォ・モリオンドオ
https://linkedin.com/in/lorenzomoriondo

On Thu, 3 Jul 2025 at 11:19, <s.mariani@unimore.it> wrote:

> Hello everybody :)
>
> I have been not particularly active in the group, but I’m noticing (also
> beyond this group) a lot of discussion around these “agent protocols” and I
> cannot help but think that there is a lot of “reinventing the wheel” that
> is happening.
>
> Coordination between agents is a long standing topic in research about
> software engineering and distributed artificial intelligence.
>
> These are just a bunch or random references that I have on the top of my
> mind while at the swimming pool, so bear with me if I am not exhaustive,
> but I think everybody interested in this new breed of “agentic systems”
> should at least be aware of.
>
> FIPA has done much to bring agent technology to production ready systems:
>  - communication language and semantics
> http://www.fipa.org/repository/aclspecs.html
>  - http specs http://www.fipa.org/specs/fipa00084/PC00084B.html
>  - coordination protocols http://www.fipa.org/repository/ips.php3
>
> A whole lot of research has also been done in multiagent systems
> coordination with alternative paradigms such as tuple based coordination,
> although perhaps with less TRL
>
> I can provide further refs and info if interested, this initial refs were
> just out of being tired of seeing over and over this way of selling
> “Agentic coordination protocols” as something new as in fact there have
> been at least 30+ years of research and dev on the topic.
>
> I go back to the shadows now :)
> Bye!
>
> Stefano Mariani, PhD
>> Tenure track researcher
>> @ Department of Sciences and Methods for Engineering – University of
>> Modena and Reggio Emilia
>> > stefano.mariani@unimore.it
>> > https://smarianimore.github.io
>> <https://www.google.com/url?q=https://smarianimore.github.io&source=gmail-imap&ust=1677865643000000&usg=AOvVaw3kYSdJbofexYY9CLIsriBn>
>>
>>
>
> Il giorno 3 lug 2025, alle ore 10:37, Joshua Cornejo <josh@marketdata.md>
> ha scritto:
>
> 
>
>
> https://www.linkedin.com/pulse/ai-agent-protocols-http-agentsa-shared-language-coordination-mtute/
>
>
>
> ___________________________________
>
> *Joshua Cornejo*
>
> *marketdata
> <https://www.google.com/url?q=https://www.marketdata.md/&source=gmail-imap&ust=1752136670000000&usg=AOvVaw3uCHYFriHbhDH0zw1zjI3S>*
>
> smart authorisation management for the AI-era
>
>

-- 
¤ acM ¤
Lorenzo
Moriondo
@lorenzogotuned
https://www.linkedin.com/in/lorenzomoriondo
https://github.com/Mec-iS