Agents — AiHummer Docs

Agents are your AI employees. On this screen you create them and set their personality, behavior, model, skills and what they are allowed to do. The screen is scoped to a workspace. Agents are shown as tiles: icon, name, model (or “default model”), status, a preview of the system prompt, and “Profile” / “Delete” buttons.

Create your first AI employee in 3 steps

If you are doing this for the first time, three steps are enough — fine-tuning can wait:

Open the “Agents” screen and click “+ Create”.
Set a name (e.g. “Support Assistant”) and a system prompt — a short description of the role: who this employee is, what it helps with and how it talks.
Click “Save”. You can leave the model empty — the default from the provider chain is then used.

What you’ll see: the new agent appears as a tile with a system-prompt preview. You can test it right away without connecting a channel: on the “Sessions” page click “+ New session” and message it.

[!TIP] Until a real model is connected, replies come from a deterministic mock — enough to confirm the agent is created and answering. Connect a provider (OpenAI, Anthropic, YandexGPT, GigaChat, Codex and more) on the Models page.

Creating an agent

The “+ Create” button opens a modal with the basic fields:

Name — required, shown in lists and channels.
Avatar — upload or pick a photo (an image).
System prompt — the text that defines the role and behavior rules.
Model — a model id; leave it empty to use the default from the provider chain.

That is enough for the agent to work. Fine-tuning happens in the profile.

Industry agent library

Next to “+ Create” there is an “Agent library” button: it opens a catalog of templates for 70 industries in nine groups — health and care, education, services and beauty, retail and commerce, property and home, internal operations, professional and industrial, personal use, enterprise. The industry list is the same as on the AiHummer website, not a second list. Template texts exist in Russian and English; any other interface language gets the English version.

Pick an industry (there is a search box) and a task: “Answers and guidance”, “Intake and preparation” or “Analysis and summaries”. The agent name is filled in automatically and can be changed.
Tick the components to add. Each one expands and shows the exact text before anything is added:
- Agent profile — short and full name, role, character, mood and form of address in separate fields;
- Instructions and task boundaries — the “Rules”, “Safety”, “Memory policy”, “Response format” and “Escalation” sections;
- Knowledge and memory lookup — a read-only tool profile (search_knowledge, memory_search, memory_get); without it the agent keeps the normal permissions of a new agent;
- a task skill — “Evidence-based answer”, “Prepare a request” or “Summarize and review”;
- Human handoff prompt — an editable prompt in the shared prompt library; unticked by default.
If a component depends on another, the panel asks you to tick that one too.
Press “Add agent”. The template contains no facts about your company: you attach documents and knowledge after adding, and every field is edited in the agent profile’s ordinary sections.

Re-submitting the same form after an error does not create a partial duplicate: the agent, skills and prompt are added in one operation. The plan’s agent limit is checked exactly as for manual creation.

Agent profile

The “Profile” button opens an editor with independently saved blocks.

Identity

Short and full name, role, how to address the user, mood, gender (unset / male / female / other), character (free text), personal phone and e-mail. These fields form the employee’s “personality” and how it introduces itself in channels.

Mailbox

Appears once the agent has a personal e-mail: IMAP and SMTP parameters (host, port, user, password). Passwords are write-only and never read back. This lets an agent work with its own inbox.

Behavior sections

A set of accordions, each with an “enable” checkbox and a text body: Known user, Rules, Safety, Capabilities, Memory policy, Refusal style, Response format, Escalation and Persona constraints.

These sections contain behavioral instructions. They do not grant tool access. Configure permissions separately in the agent’s dedicated controls.

Shared blocks

You can link company-wide shared blocks from the workspace catalog (a single set of company rules) — see Skills & prompts. A block shows a sensitivity badge.

Skills

The agent’s own skills (create/edit/delete: name, description, body, enabled, order; the slug is generated by the server) plus linking the workspace’s shared skills.

Skills are instructions. Creating, editing, linking, importing or combining skills alone grants no additional tool access and does not switch execution mode. Task execution is configured separately: ordinary work is the default; the agent’s background stages with verification require an explicit selection. The mode adds no permissions. An execution policy included in an import is a separate, explicitly selected component.

Task execution: Ordinary work / Long running with verification.

Setting rights and model

Rights/permissions — Configure tool access and execution permissions separately from prompts and skills. Existing roles, Approvals and Moderation still apply.
Model can be set at creation (free text) or left empty for the default; which models are allowed at all is on Instructions → allowed models.

If it didn’t work

The agent replies in a generic/templated way — the mock model is probably still active. Connect a real provider on the Models page.
I don’t see the agent in a channel — creating an agent isn’t enough; you must bind it to the channel in the “Bindings” block on the Channels screen.
I set a model but the agent errors out — that model may not be in the allowed list, or the provider’s Test didn’t pass; check Models and allowed models.
I want to limit what the agent may do — use the agent’s separate tool-access and execution-policy settings. Editing instructions alone does not grant or revoke technical access. See Approvals.

Tips

Keep the system prompt short and push details into sections and shared blocks — they are easier to reuse across agents.
Sensitive shared blocks are edited through the approvals queue.

Orchestration — wire several agents into a graph.
Skills & prompts — reusable blocks.
Models — providers and priority.