ZDNET’s key takeaways
- Askimo is a flexible, all-in-one AI tool.
- The plans feature alone makes Askimo worth using.
- You can connect Askimo with cloud or local AI.
I’ve been hunting for a one-stop-shop AI tool that can handle most of the tasks I need, and I may have finally found it.
That tool is called Askimo, and it’s available for Linux, MacOS, and Windows. It’s free (as long as you use local AI, which is what I do), and it includes features like projects, agents and skills, resource collections, plans, the ability to connect to local AI, MCP servers, voice input, notifications, keyboard shortcuts, web search, RAG configuration, and more.
Also: How I built the writing app I needed in 2 hours for free
I’ve used Askimo for general AI chat, creating CVs and cover letters, and early stage app development. I’ve found Askimo to be very flexible, as long as I’m willing to do a little extra work.
Extra work? Seriously?
OK, it’s not as bad as you think (at least for some functions), and the reason I do a little extra work is that I stick with local AI. If you use Askimo with a third-party AI service, you’ll find that it does a lot of what I have to do manually without the extra effort. The extra work is isolated to the agents and projects feature, where you may have to copy-and-paste the content of, for example, a resume into a chat, because local AI does not have the ability to access folders (which is strange to me, but it’s how Askimo works), or connect the tool to a third-party AI service.
What I have found though is that Askimo aligns more than I tend to work. I want to be part of the process, not have the technology do everything for me.
How I set up Askimo
I’ll walk you through how I built a project to create a cover letter based on a tech-writing resume for the freelance writer position at Company X.
Also: How to feed my files to a local AI for better, more relevant answers
Before I could do that, I first had to connect Askimo to my locally installed Ollama AI instance configured in settings. After opening settings, I added a new AI provider. I first had to run a working instance of Ollama, with the models I wanted to use drawn, because Askimo can’t draw models automatically.
For example, if I wanted to use the model llama3.3:3b, I would draw it with Ollama using the command:
ollama pull llama3.2:3bAfter that it will be available to Askimo.
Within the Add AI Provider window, I selected Ollama and then clicked Connect.
In the resulting window, all I had to do was click Next.
I then had to select the model I wanted to use and click save. That’s it. Askimo is now ready to use Ollama.
Next, projects: a little frustration
Now that Askimo is connected to Ollama, it’s time to create a project. Let me warn you first: projects can be a little frustrating because you have to do a little extra work.
To create a project, I clicked on Projects in the left sidebar and then clicked New Project.
Jack Wallen/ZDNETIn the Create New Project pop-up, I gave it a name (Cover Letters), clicked Add Source and selected Resume File. Next, I clicked on the Default Directory dropdown, selected Formal Writing and clicked Save.
Now that the file was added to the project, all I had to do was select the project from the project list, and then type what I wanted in the chat, like:
Write a cover letter for the position of freelance writer for Company X using the contents of the resume added to the RAG sourcesUnfortunately, Axios failed in this task.
Also: Stop local AI from crashing your Linux PC – 4 tips
Here is what I discovered. Due to permission issues that prevent Askimo from reading the files on your drive, your best bet is to attach the file to the chat, whether it’s a standard chat or a Projects chat. This is frustrating because projects cannot use Resource Attachments Gathering (RAG) sources because they cannot access or process their content. When I asked why Askimo could not read the contents of the files added to RAG sources, it responded with three reasons:
“Files can be extremely large, which can cause technical problems during recovery; some files contain sensitive information that cannot be accessed or shared for security or privacy reasons, or file formats cannot be optimized for text-based interaction.”
What Askimo neglected to tell me was that my local model, Ollama, does not have access to the contents of the files. To do that, I need to use an API key for a third-party AI service.
Fortunately, there is an even better tool within Askimo, one that comes in very handy and avoids paying for an API key.
Plans are my favorite
Plans are one of my favorite features in Askimo. Plans are essentially AI-powered, multi-step workflows that you can create and use whenever you need them. Plans prevent you from creating and re-creating long, complicated chats that take extensive time and effort and confuse the AI in use.
As the name suggests, plans allow you to divide work into steps that run, then pass the results of each step on to the next.
Also: Someone made a free Photoshop clone – so I had to try it on Linux
You will find six pre-built plans in Askimo: Blog Post Writer, Job Application Writer, Research Report, Competitor Analysis, Meeting Notes Processor, and System Design.
Let’s use the Job Application Writer plan to do the same thing we did before, only this time we’ll find it generates results much faster.
To do this task, I clicked Plans in the left sidebar and then selected Job Application Writer. In the resulting window, the first thing I did was to select a good model for this work (such as llama3.2:3b). When I did that, I did the following:
- Add an input for job description (which can be taken directly from the job listing).
- Background: Paste my CV summary or resume
- Add the name of the company
I jumped on LinkedIn, found a job to use as an example, filled in the necessary information and clicked Run Plan. Askimo set out to do his thing, and I watched it. In less than two minutes, I had a professional resume that I could then use for follow-up questions, such as writing a cover letter for this position (which I did, and it turned out very well).
You can also create your own plans written in YAML. Yes, they are complicated at first, but once you get the hang of writing them, the sky is the limit. The YAML files are laid out like this:
inputs:
- key: topic # referenced in steps as {{topic}}
label: Topic # caption shown in the UI
type: text # text | multiline | toggle | number
required: true # plan will not run if blank
default: "" # pre-filled value
hint: Enter a topic # placeholder / helper textYou create sections for inputs, steps and workflow. Your best bet is to select a pre-built plan, open the three-dot menu, and then select copy and edit. Then the Plan Editor will open so you can see how they are laid out. You can edit a copy and use it as a template to build your own plans.
Plans is the game changing feature that I plan to use quite a bit.
My conclusion about Askimo
Askimo is one of those tools that can do a lot more than you think, but it takes a little effort to understand how it all works. The beauty of this app is that the second you learn one thing about it, it opens more doors to other features and aspects.
Even if I only used the plan feature, it would allow me to create automated workflows for almost anything. This feature alone is worth the time and effort to get up to speed with this tool.
I will also continue to work on trying to make it possible to use projects so that Askimo can actually read the contents of RAG files so that they can be used as input sources. Until then, I’ll continue to use the chat and plans feature and not have to rely on more than one app for anything AI-related (except maybe Vibe coding, which VS Code does very well).
See full bio