Agent concepts

An agent is the unit Tale reaches for when the same question is going to come back. It is the four-knob combination of instructions, knowledge, tools, and a model — the four things you change to make the agent behave differently. Editors and Developers build them; Members and other roles run them.

This page hands you the mental model the rest of the section assumes. Read it once before you build your first agent; come back to it when you can't remember whether a behaviour you want to change lives in the instructions, the knowledge, the tools, or the model.

The four knobs

Instructions are the system prompt — the prose that frames every reply. Keep instructions short, opinionated, and concrete; long instructions get diluted in long conversations. Specify the voice, the constraints, and the refusal cases.

Knowledge is what the agent can reference. Bind documents, customers, products, vendors, or websites from the knowledge base; the agent retrieves chunks at reply time and cites them. Knowledge that is not bound is invisible to the agent — there is no implicit pull from the org's whole library.

Tools are what the agent can do beyond reply with text. Built-in tool families cover web, files, RAG over knowledge, code execution, sub-agent delegation, workflow invocation, MCP servers, and human input. Toggle them per agent — every tool you grant widens the trust boundary, so keep the list short.

Model is the LLM behind every reply. Pick the primary, set a fallback, and Tale resolves at request time. Switching the model does not re-train anything — the agent's other three knobs are the model's "memory" of the job.

Skills as a bundle

A Skill packages instructions and (optionally) a sandbox script into a reusable bundle you can attach to an agent. Reach for a skill when the same pattern appears across multiple agents — a writing voice, a calculation, a multi-step task. Skills compose with the four knobs: an agent with three skills has the instructions of each skill plus its own.

The Skills concept page covers the trade-off between Skills and inline instructions in detail: see Agent skills.

Putting it together — a support-triage agent

A first useful agent is the support-triage one: it reads the inbound conversation, decides whether to answer directly, escalate to a human, or hand off to a specialist. The four knobs:

• Instructions: a one-paragraph voice + three explicit refusal cases.
• Knowledge: the product documentation and the FAQ folder; not the source code.
• Tools: RAG, web search, and the sub-agent tool for escalation. No code execution.
• Model: a capable model for the primary, a smaller one for the fallback when the primary is rate-limited.

The conversation then flows: user message → instructions frame the reply → knowledge retrieval finds three relevant chunks → tools either answer or delegate → the reply lands with citations.

When to reach for it

A single agent is the right shape when the conversation stays in one domain and one voice. Reach for an automation when the work is multi-step and you want approvals or scheduling in between; reach for a raw chat (no agent) when you are exploring an answer yourself and the model's defaults are fine.

Build one

The four knobs are what every Tale agent is made of: change one of them and you have changed the agent's behaviour, change three and you have made a new product. The natural next read is Build your first agent — it walks the four knobs end to end on a fresh instance.