Use your Mind

Understand skills, agents, automations and loops

Your Mind can hold instructions for your AI tool and run installed programs with agreed access. A skill is a written procedure. An installed agent is a program that does a defined job. An automation joins available actions into a triggered sequence. A loop adds a lasting case, checks, branches and waits. This page helps you understand which part is doing the work and what you need to check before relying on it.

Content updated .

Contents

Follow a skill in your connected tool

Your connected agent is the AI tool you use to work with your Mind. It can discover and read skills, then follow the procedure with the tools and permissions available in that client. Reading a skill does not execute its steps or give it more access. Use or create a skill explains discovery and tested authoring.

Install a program with its own access

An installed agent has its own identity and a declared job, information scope, model budget and any required service connections. It can run on the computer hosting the Mind without your chat remaining open. Installation, a scheduled attempt and a useful output are separate results. Install an approved agent explains the handoffs.

An app can also supply an action that another workflow may call. The live action list describes what is installed, the inputs each action accepts and its effects. A catalog listing alone does not make an action available to your workflow.

Join actions into an automation

An automation has one trigger and an ordered chain of installed actions. It is useful when the steps follow a clear sequence. A workflow that must pause for a person or judge a result belongs in a loop. Automate a routine task walks through the choice and review.

Follow recurring work as cases

A loop is a recurring workflow with checks, branches and possible waits. Each occurrence creates a case that records its progress. Approval accepts the definition and starts its scheduled runs in shadow mode; activation separately lets it make real changes within its autonomy limits; actions above those limits still wait for the owner. A shadow run performs real reads and checks while holding back any action that would change something. Your first loop explains setup, and loop cases explains individual results.

Check access, effects and spending separately

Decide which records a program may use, which services it may call and what requires a person. Model routes, app budgets and loop case limits control different things. Autonomy, budgets and access explains them; Mind settings points to the relevant controls.

A service connection supplies an authorised route to an outside system. It does not make every action read-only or enable scheduled capture by itself. Check the selected app's purpose and setup in Apps, sensors and connectors.

Related

Choose the right kind of help · Use a skill or turn a repeated job into one · Install an approved agent and check its first result · Automate a routine task with installed actions · Build your first recurring workflow with a loop

Practical guides for people and agents

Start with one useful job

Join the waitlist

Plain text for AI agents. Same content as the human page.

## For agents

- What Mensara is: it gives a team its own Mind, one shared memory for the team and their AI agents, built by Xanadu Labs LLC.
- Status: signup is not open. A person joins the waitlist at https://www.mensara.ai/#waitlist. Agents cannot create an account or sign up.
- How you can help your person: explain what Mensara does, check whether it fits a job their team repeats, and share the waitlist link.
- Agent references: https://www.mensara.ai/llms.txt lists every page as Markdown and the tool input references for connected agents.

---

# Understand skills, agents, automations and loops

Content updated 29 October 2026.

How Mensara works · Use your Mind · [How it works](https://www.mensara.ai/how-it-works)

Your Mind can hold instructions for your AI tool and run installed programs with agreed access. A skill is a written procedure. An installed agent is a program that does a defined job. An automation joins available actions into a triggered sequence. A loop adds a lasting case, checks, branches and waits. This page helps you understand which part is doing the work and what you need to check before relying on it.

[Page as Markdown](https://www.mensara.ai/how-it-works/skills-apps-automations.md)

## Follow a skill in your connected tool

Your connected agent is the AI tool you use to work with your Mind. It can discover and read skills, then follow the procedure with the tools and permissions available in that client. Reading a skill does not execute its steps or give it more access. [Use or create a skill](https://www.mensara.ai/how-it-works/use-and-create-skills) explains discovery and tested authoring.

## Install a program with its own access

An installed agent has its own identity and a declared job, information scope, model budget and any required service connections. It can run on the computer hosting the Mind without your chat remaining open. Installation, a scheduled attempt and a useful output are separate results. [Install an approved agent](https://www.mensara.ai/how-it-works/install-an-agent) explains the handoffs.

An app can also supply an action that another workflow may call. The live action list describes what is installed, the inputs each action accepts and its effects. A catalog listing alone does not make an action available to your workflow.

## Join actions into an automation

An automation has one trigger and an ordered chain of installed actions. It is useful when the steps follow a clear sequence. A workflow that must pause for a person or judge a result belongs in a loop. [Automate a routine task](https://www.mensara.ai/how-it-works/automate-a-task) walks through the choice and review.

## Follow recurring work as cases

A loop is a recurring workflow with checks, branches and possible waits. Each occurrence creates a case that records its progress. Approval accepts the definition and starts its scheduled runs in shadow mode; activation separately lets it make real changes within its autonomy limits; actions above those limits still wait for the owner. A shadow run performs real reads and checks while holding back any action that would change something. [Your first loop](https://www.mensara.ai/how-it-works/first-loop) explains setup, and [loop cases](https://www.mensara.ai/how-it-works/loop-cases) explains individual results.

## Check access, effects and spending separately

Decide which records a program may use, which services it may call and what requires a person. Model routes, app budgets and loop case limits control different things. [Autonomy, budgets and access](https://www.mensara.ai/how-it-works/control-agent-autonomy) explains them; [Mind settings](https://www.mensara.ai/how-it-works/mind-settings) points to the relevant controls.

A service connection supplies an authorised route to an outside system. It does not make every action read-only or enable scheduled capture by itself. Check the selected app's purpose and setup in [Apps, sensors and connectors](https://www.mensara.ai/how-it-works/apps-sensors-connectors).

## Related

- [Choose the right kind of help](https://www.mensara.ai/how-it-works/right-kind-of-help)
- [Use a skill or turn a repeated job into one](https://www.mensara.ai/how-it-works/use-and-create-skills)
- [Install an approved agent and check its first result](https://www.mensara.ai/how-it-works/install-an-agent)
- [Automate a routine task with installed actions](https://www.mensara.ai/how-it-works/automate-a-task)
- [Build your first recurring workflow with a loop](https://www.mensara.ai/how-it-works/first-loop)

Previous page: [Bring email, calendar events and files into your Mind](https://www.mensara.ai/how-it-works/connections-and-capture)

Next page: [Review changes and check what was saved](https://www.mensara.ai/how-it-works/changes-approvals-receipts)

A proposal is a suggested change to information saved in your Mensara Mind.

[Join the waitlist](https://www.mensara.ai/?guide_attribution=eyJfcmFpbHMiOnsiZGF0YSI6eyJvcmlnaW5fcGF0aCI6Ii9ob3ctaXQtd29ya3Mvc2tpbGxzLWFwcHMtYXV0b21hdGlvbnMiLCJjdGFfaWQiOiJtYXJrZG93biIsImNvbnRlbnRfcmV2aXNpb24iOiI2ZmQzNTliYmNiMWI2NDA4In0sInB1ciI6Imd1aWRlLWNsaWNrIn19--f054f82db425e008a375ac0050a72013d0fd467e#waitlist)