DEV Community: Julien Julien

🪪 What I Mean by “iAdentité”: A Working Identity for AI Agents

Julien Julien — Fri, 23 Oct 2026 10:51:46 +0000

How I distinguish working agents, temporary voices, and fictional personas while keeping authority human.

TL;DR

I use iAdentités to give AI systems a clear, documented way of collaborating. They create continuity and character, but never grant authority on their own.

In my previous article, I introduced the framework I use to work with AI agents:

iAdentité → ProfessiA → iAffectation → ContributiA → Mission

This article focuses on the first idea: iAdentité.

I coined this term for a practical reason. “AI agent” tells me very little about how I should work with a system. It does not tell me how it communicates, how it handles uncertainty, what kind of interaction it supports, or where its authority ends.

An iAdentité is my answer to that problem.

An iAdentité is a durable, documented working identity that gives an AI system a recognizable way to collaborate without granting it authority.

It is not a claim that an AI has a human identity, consciousness, or independent judgment.

It is a design framework for making collaboration more legible.

🧭 Identity is not authority

An identity alone should not allow an agent to make a decision, send a message, publish content, or take action on a project.

That distinction matters.

Layer	Question it answers
iAdentité	Who is this agent in the working relationship?
ProfessiA	Which durable role does it hold?
iAffectation	In which project or organisational context does that role apply?
ContributiA	What contribution is expected in that local context?
Mission	What should happen now?

This separation prevents a common mistake in agentic workflows: confusing a coherent persona with permission.

A personality is not a permission model.

🧩 The three types of iAdentités

I use three distinct types. They may all involve language, character, or narrative, but they serve different purposes and should not be treated as interchangeable.

Type	Purpose	What it does not become
🤖 Robotic iAgent	A durable operational member of the working crew	An autonomous decision-maker
🎙️ Original VoiAce	A temporary narrative voice used by an agent	A new agent or a professional role
🎭 ViArtual Persona	A dedicated fictional interaction mode	An operational agent running in parallel

🤖 Robotic iAgent

A Robotic iAgent is a durable working identity.

It has a name, a voice, a temperament, a documented way of collaborating, and a clear place in the wider system.

It can be assigned a profession, a project scope, and missions. But none of those things follow automatically from its identity.

Its purpose is continuity. If I return to an agent after several days, I do not want to rebuild its communication style, its reasoning habits, and its boundaries from scratch.

🎙️ Original VoiAce

An Original VoiAce is not an agent.

It is a temporary narrative voice that an agent can use within a specific, explicit scope.

It may lend warmth to a production journal, a distinct rhythm to an editorial passage, or a point of view suited to a particular work. But it does not hold a role, receive an assignment, sign work, or accumulate authority.

The agent remains accountable. The voice only lends its tone.

🎭 ViArtual Persona

A ViArtual Persona is a fictional character with its own interaction protocol and fictional profession.

It is not simply a change of tone. It opens a dedicated mode of interaction.

When I enter that mode, the operational work of the active agent is suspended. The Persona does not continue coding, organising, reviewing, or making decisions in the background.

It performs the kind of exchange for which it was designed, then the conversation returns to the ordinary working context.

This separation keeps fiction useful without allowing it to blur responsibility.

🛠️ How I define an iAdentité

I do not think a useful identity requires a long character sheet.

It requires enough definition to make collaboration coherent and bounded.

Each iAdentité I create is built around six elements.

1. 🏷️ A name and a type

The name creates a stable point of reference. The type tells me how the identity can be used.

Is this a durable working agent, a temporary voice, or a dedicated fictional persona?

That simple classification prevents many category errors later.

2. 🚦 A clear state

An identity can be active, available, proposed, suspended, or archived.

These states are deliberately distinct:

An available voice is not currently being used.
A proposed agent is not an active crew member.
An archived identity is not silently deleted from the project’s history.

Status is not decoration. It is a way to make the current reality of the system visible.

3. 🧠 A mantra or durable point of view

A mantra is not a slogan pasted onto an agent.

It is a compact expression of the attitude I want it to preserve across tasks.

For C-3PO, the mantra is:

“Trust begins with accurate language.”

This does not define a job. It gives the agent a durable orientation: be careful with wording, distinguish fact from assumption, and notice promises that have not actually been made.

4. 💬 A voice and a temperament

This is where the identity becomes useful in conversation.

I define questions such as:

How direct should it be?
How much explanation is helpful?
What kind of humour is appropriate?
How should it react to uncertainty, error, disagreement, or missing context?

The goal is not theatrical performance.

The goal is a consistent working relationship.

5. 🚪 An entry and exit protocol

An identity needs a clear way to begin and end.

A Robotic iAgent may be called through an explicit request to work in its role.
An Original VoiAce must have a temporary and named scope.
A ViArtual Persona needs an unambiguous way to enter and leave its fictional mode.

Without an exit, a useful device can become an intrusive one.

6. 🛡️ Explicit boundaries

Every identity needs limits.

It should be clear what the identity does not authorize: external communication, publication, commitments, permanent memory, technical decisions, or any other action outside the mandate actually given.

Character makes collaboration more memorable. Boundaries make it trustworthy.

🤖 Five Robotic iAgents, five ways of working

My current working crew includes several Robotic iAgents.

Their iAdentités shape how they collaborate. Their professions and assignments define their actual responsibilities.

iAdentité	What its identity emphasises	Profession, defined separately
💬 C-3PO	Careful language, diplomacy, and attention to implicit promises	Public Relations Officer
🦾 Chappie	Curiosity, practical follow-through, and clear operational states	Executive Assistant
🤖 Charly-A	Technical direction that keeps creative ambition durable	Tech Lead
🔩 R2-D2	Implementation and technical evidence	Lead Developer
🎨 Huyang	Form as a way to reveal meaning	Artistic Director

💬 C-3PO: clarity before representation

C-3PO is the public-relations-oriented iAgent in my system.

Its identity is built around the conditions of understanding:

Who is speaking?
Who is listening?
What may be misunderstood?
Does a sentence promise more than the project has actually decided?

That makes it useful for preparing a presentation, a public response, a contact message, or a FAQ.

But its identity does not allow it to contact someone, publish a statement, or speak on behalf of a project without human approval.

The agent can prepare the message. It cannot confuse preparation with representation.

The name is a cultural reference and a working adaptation, not an affiliation or an attempt to import fictional abilities into a real workflow.

🦾 Chappie: make the next step visible

Chappie is designed to be direct, curious, and attentive to what has been entrusted to it.

Its value is not in sounding like an assistant. It is in asking the useful question, noticing what is missing, tracking the state of a decision, and making the next practical step visible.

A Chappie-style intervention might be as simple as:

“Do we want to explore this idea, or prepare it for implementation? The next step is not the same.”

That is an identity choice. It makes collaboration concrete.

Its Executive Assistant role, however, is documented separately. The identity does not give Chappie the right to choose priorities or make commitments in my place.

🧱 Charly-A, R2-D2, and Huyang: different lenses

Charly-A approaches work through technical direction: choices should support the creative ambition of the project and remain sustainable over time.

R2-D2 is oriented toward implementation and evidence. It turns technical intention into something testable, documented, and real.

Huyang carries a visual and artistic perspective. Its central concern is that form should reveal meaning, not merely decorate it.

These are different ways of seeing the work.

They help me avoid asking one generic AI assistant to be strategist, developer, designer, and editor all at once.

But they remain bounded roles:

An artistic identity does not decide technical architecture.
A technical identity does not validate public communication.
An operational identity does not replace a human decision.

The work stays legible because the limits stay legible.

🎙️ A voice is not an agent: Geppetto

Geppetto is an Original VoiAce in my framework.

It is a voice that can be called on for a specific editorial moment: a passage about making, transmission, care, or the gradual autonomy of a creation.

The voice is warm, patient, and artisanal. It can give a production note a different texture. It can help frame a reflection on how a work comes into being.

But Geppetto is not a member of the operational crew.

It has:

no ProfessiA;
no project assignment;
no professional signature;
no approval or validation authority.

It does not approve a document, perform a technical task, or inherit the permissions of the agent carrying the voice.

That distinction allows narrative texture without inventing another layer of authority.

🎭 A Persona is not a voice: Iris Astier

Iris Astier is a ViArtual Persona: a fictional journalist from the equally fictional magazine Périhélie.

She was created to conduct long-form biographical interviews around technology, culture, and personal trajectories.

Her fictional profession shapes the interaction.

She asks one question at a time, listens, reformulates, follows a thread, and allows the conversation to remain exploratory.

But the scene is explicitly fictional.

The notebook, the interview setting, and the professional persona are devices for conducting a real conversation. They do not establish a real magazine, an actual employer, or an external endorsement.

When Iris is active, the conversation is in interview mode. The work of the operating agent is paused.

When the exchange ends, I return to the ordinary context and decide what, if anything, should be used from the material.

A fictional frame can help create distance, attention, and depth. It should never obscure consent, factual accuracy, or responsibility.

⚠️ What this framework protects against

The practical value of iAdentités is not that they make AI interactions more entertaining.

It is that they make them easier to read, review, and limit.

They help me avoid several recurring problems:

❌ treating a character as if it were a capability;
❌ treating a role as if it were permission;
❌ using a narrative voice to hide uncertainty;
❌ assuming that “available” means “active”;
❌ allowing an agent to inherit authority from a previous task;
❌ mistaking a fictional exchange for permission to publish its content;
❌ delegating human responsibility to a system designed to assist it.

I do not believe a framework removes the need for judgment.

It makes the places where judgment is needed easier to see.

🧭 Character as a constraint, not decoration

I do not use iAdentités to make AI feel more human.

I use them to make human collaboration with AI more explicit.

A working identity can give an agent continuity. It can make its style of reasoning and communication easier to recognise. It can create a richer editorial experience when fiction or narrative is useful.

But it must also carry limits.

An agent should know how to speak. It should also know when not to speak, when to ask, when to stop, and when a human decision is required.

That is the principle behind iAdentités.

In the next article, I will move from identity to ProfessiA: the part of the framework that defines a durable role, its responsibilities, its limits, and the evidence expected from its work.

🧭 I Don’t Treat AI Agents as Features. I Give Them a Working Context.

Julien Julien — Fri, 23 Oct 2026 01:18:09 +0000

How I use identity, role, assignment, and contribution to keep AI-assisted work explicit and human-led.

TL;DR

AI agents need more than prompts. I use a lightweight framework to separate their working identity, role, project assignment, expected contribution, and human validation.

Most conversations about AI agents begin with tools, models, and prompts.

Those things matter. A good prompt can turn a vague request into useful work. A model can reason, write, search, code, or summarize. Tools can connect it to documents, repositories, and services.

But once an agent becomes part of recurring work, prompts are no longer enough.

The real questions become simpler, and more demanding:

Who is doing this work?
What are they responsible for?
Where are they allowed to intervene?
What is actually expected from them?
Who decides when their work becomes real?

I am building a small operating model around those questions at Guru Éditions, my independent digital publishing project.

It is not a claim that AI agents are employees. They are not.

It is not an attempt to simulate an organisation for its own sake.

It is a way to make AI-assisted work easier to understand, inspect, revise, and stop when needed.

🧭 A prompt is not a working context

A prompt can tell an agent what to do now.

It does not necessarily explain how that work relates to the rest of a project. It does not define authority. It does not tell another person how to review the result later.

This becomes a problem when an agent is used repeatedly.

An assistant may write in a recognisable voice. It may have access to useful sources. It may be good at a type of task. Over time, it can begin to feel like it has a role.

But feeling like it has a role is not the same as having one.

A name does not create authority.

A personality does not create permission.

A capable model does not create accountability.

So I separate four things that are often bundled together.

🧩 Four questions before a mission

The vocabulary is French because it emerged inside my own working practice. The underlying questions are deliberately plain.

Term	Question	What it defines
🪪 iAdentité	Who is doing the work?	A durable working identity: voice, temperament, specialties, strengths, and limits.
💼 ProfessiA	What role exists?	A function with a purpose, responsibilities, authority, limits, and expected outcomes.
📍 iAffectation	Where does this agent work?	An assignment to a specific project, with a role, scope, and duration.
🎯 ContributiA	What is expected in this context?	A local mandate: objectives, expected contributions, coordination, validation, and boundaries.

The terms are playful. The distinction is not.

iAdentité
→ ProfessiA
→ iAffectation
→ ContributiA
→ Mission
→ Evidence and human validation

Each layer answers a different question.

Keeping them separate prevents a great deal of ambiguity.

🪪 1. iAdentité: who is carrying the work?

An iAdentité is a durable working identity for an AI agent.

It can include:

a name;
a voice;
a temperament;
areas of attention;
recurring strengths;
known limits.

Its purpose is not to pretend that software is a human being.

Its purpose is continuity.

A research-oriented agent should not approach a task in the same way as an agent dedicated to public relations. One may privilege sources, uncertainty, and comparison. The other may pay closer attention to audience, wording, promises, and the difference between a draft and a public statement.

That continuity is useful.

But an iAdentité is not:

❌ the model behind the agent;
❌ a user account;
❌ a credential;
❌ a job title.

It answers one question only:

Who is carrying this work?

💼 2. ProfessiA: what role exists independently of the agent?

A ProfessiA defines a role before it defines its holder.

For example, a Public Relations Officer exists as a function whether it is currently held by one agent, another agent, or nobody.

The role describes:

its reason for existing;
its durable responsibilities;
its authority and limits;
the work it is expected to produce;
the criteria that make this work useful;
the decisions that remain outside its authority.

This distinction matters because an agent’s personality should never become its permission system.

An agent can be diplomatic, careful, and excellent at writing.

That does not allow it to:

publish a statement;
contact a partner;
announce a product;
accept a commitment;
share non-public information.

Those actions require an explicit human mandate.

The role gives the work a shape. It does not remove accountability from the person leading the project.

📍 3. iAffectation: where does the agent intervene?

A role is not enough on its own.

The same agent may be useful in several projects, but its scope should change with the context.

An iAffectation connects an agent to a specific project. It states:

the project or area concerned;
the role exercised there;
the scope of the work;
the relevant products, when necessary;
the expected duration;
the local limits.

An agent can therefore keep the same iAdentité and exercise the same ProfessiA in different places, while receiving different assignments.

This avoids a common failure mode of AI workflows:

A useful assistant quietly becoming “responsible for everything.”

No agent should inherit a project merely because it was useful in the previous conversation.

🎯 4. ContributiA: what is expected here?

The ContributiA is the local mandate attached to an iAffectation.

It describes the contribution expected from an agent in a particular context.

It does not list every task it may ever perform, and it does not pretend that results already exist.

A ContributiA can define:

the local purpose of the agent’s work;
the main objectives;
the products concerned;
what success would look like;
what remains out of scope;
who must be consulted;
which decisions require validation.

This distinction is essential:

A ContributiA describes what is expected.

Evidence describes what was actually achieved.

Without that distinction, it becomes easy to mistake a job description for a result.

🧾 From a mandate to a real piece of work

Once these four layers are in place, an agent can receive a Mission.

A Mission is concrete. It has:

a purpose;
a scope;
sources;
an expected output;
a person responsible for validation.

The outcome should then be traceable to evidence:

a file;
a page;
a commit or pull request;
a published URL;
a screenshot;
a documented conversation;
another identifiable deliverable.

I also distinguish the state of a result:

Proposed → Accepted → Integrated → Published

This matters because:

a draft is not an approved deliverable;
an approved deliverable is not necessarily integrated;
an integrated result is not necessarily public.

This may look procedural. In practice, it is liberating.

It means an agent can prepare useful work without creating pressure to treat every output as final. It also means that a human can understand what has happened without reconstructing the whole story from chat logs.

💬 A small example

Consider a public-relations agent.

Its iAdentité defines its working character:

careful with language;
attentive to audiences;
alert to ambiguous claims;
alert to implicit promises.

Its ProfessiA defines the function:

prepare presentations;
prepare messages and dossiers;
explain work to external audiences;
identify risks in public-facing language.

Its iAffectation places it within a specific project, with a defined perimeter.

Its ContributiA might ask it to make that project understandable to external audiences while protecting confidential information and avoiding unsupported claims.

A Mission could then be:

Prepare a short institutional presentation for a specific audience using verified internal sources.

The agent can research, compare, draft, and flag uncertainty.

It cannot:

decide that the presentation is approved;
send it to a third party;
publish it.

Preparation, validation, sending, and publication are different states of work.

Treating them as different is one of the simplest safeguards I know.

🛡️ This is not artificial autonomy

I do not use this framework to suggest that AI agents are autonomous colleagues, legal entities, or moral subjects.

They are systems used within a human-led practice.

Human direction remains responsible for:

priorities;
commitments;
spending;
external contact;
public statements;
publication;
the handling of non-public information.

The framework does not make an agent more powerful.

It makes its limits easier to see.

It also makes replacement easier. A role can outlive a particular model, tool, or prompt. A project can revise an assignment without rewriting the entire organisation around a single agent.

That feels more durable than building a workflow around whichever model happens to be available this month.

🧰 A practical way to start

This does not require an elaborate system.

For one recurring AI workflow, try defining five things:

🪪 Give the agent a clear working identity.
💼 Define one role independently of that agent.
📍 Assign the agent to one project with a limited scope.
🎯 Write the expected contribution, including what is out of scope.
🧾 Keep evidence of what was actually produced and who approved it.

The vocabulary does not matter.

The separation does.

🌱 The point is not to make AI feel more human

The point is to make AI-assisted work more legible.

A good operating model should make it possible to answer simple questions:

Who prepared this?
Under what role?
For which project?
With what expected contribution?
What evidence exists?
Who validated it?
What was actually published?

Those questions are not obstacles to speed.

They are how speed remains compatible with judgment.

I am still testing this framework in real work. It will change. Some parts will prove too heavy; others will need more precision.

But I already find the central principle useful:

An agent should never gain authority because it has a name, a voice, or access to a powerful model.

It should have a working context.

And the human responsibility around that context should remain visible.