microsoft / microsoft/PowerToys

Command Palette: Installed apps plugin returns overly broad results; add search fuzziness control

Open
#45,653 0 comments 0 reactions 0 assignees View on GitHub
Needs-Triage Product-Command Palette
Dominant language
C
Stars
139k
Forks
8.6k
PR merge metrics
PR metrics pending

Description

### Description of the new feature / enhancement

When searching for an application like "ollama", PowerToys Run returns relevant entries: the installed application and a web search suggestion. In contrast, Command Palette's installed apps plugin returns many unrelated items (e.g., PowerShell, Visual Studio tools, etc.), as shown in the attached screenshot.

The search algorithm appears to be too broad, matching partial terms or keywords in a way that dilutes precision. I suggest:

Refining the relevance algorithm for installed applications to prioritize exact or near‑exact matches.

Adding a user‑configurable setting for search fuzziness (e.g., strict, normal, broad) so power users can tailor the behavior to their preference.

### Scenario when this would be used?

When quickly launching an application, users expect the top results to be the most relevant. An overly broad search forces extra scrolling or typing, defeating the purpose of a launcher. This is especially important for power users who rely on muscle memory and expect consistent, precise results.

### Supporting information

Attached is a screenshot showing the results for "ollama" in Command Palette (many irrelevant entries) compared to the expected behavior from PowerToys Run. The ability to adjust search fuzziness would give users control over the trade‑off between recall and precision.

Image

Image

Contributor guide

Open the contributing guide

Research direction

Start by locating Command Palette's installed apps plugin and the search relevance code; the issue does not name files or tests. Compare the reported "ollama" results with the expected precise behavior, then define completion as more relevant ranking plus a user-configurable strict, normal, or broad fuzziness setting.

Written by the indexing model from the issue text.

Assessment

Domain
desktop, search
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
42/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.