cd /news/ai-tools/show-hn-mira-search-for-files-semant… · home topics ai-tools article
[ARTICLE · art-17014] src=github.com pub= topic=ai-tools verified=true sentiment=↑ positive

Show HN: Mira – Search for files semantically – no exact filenames required

A developer released Mira, a local-first desktop application that enables users to search files using plain English without requiring exact filenames. The app indexes folders locally, extracting metadata and text from PDFs and documents while generating semantic embeddings for meaning-based retrieval, with an optional Gemini API provider for cloud processing. Mira is available for macOS via GitHub Releases, though users must remove a quarantine flag on first launch due to the app not being notarized by Apple.

read1 min publishedMay 28, 2026

From Latinmīrāre-to look, to see, to behold.

Mira is a local-first desktop application that turns your folders into a searchable, intelligent workspace. Search files in plain English - no exact filenames required.

Download the latest macOS build from the GitHub Releases page.

Open the

.dmg

and dragMira intoApplications

. - Because Mira is not yet notarized by Apple, macOS will show

*"Mira is damaged and can't be opened"*on first launch. Remove the quarantine flag by running this once in Terminal:

xattr -cr /Applications/Mira.app

Launch Mira from

Applications

.

mira_compressed.mp4 #

Metadata search- instantly query file names, types, sizes, and dates across indexed folders** Content extraction**- full-text search inside PDFs, documents, and text files** Semantic search**- meaning-based retrieval powered by local embeddings by default, with an optional Gemini API provider** Live indexing**- add a source folder and Mira begins indexing in the background immediately** File preview**- open or reveal any result directly from the app

Add a source- point Mira at any folder on your machine** Index**- Mira walks the directory tree, records file metadata, extracts text from supported formats, and generates semantic embeddingsSearch- queries are matched against metadata, extracted text, and semantic meaning simultaneously, then ranked by relevance

By default, processing happens locally and your files never leave your device. If you enable the optional Gemini API provider in settings, Mira sends the content needed for embeddings to Google's API using the key you provide.

Mira - look deeper into your files.

Prerequisites: Rust, Bun, and the Tauri v2 system dependencies for your OS.

bun install

bun run tauri dev

bun run tauri build

The distributable will be in src-tauri/target/release/bundle/

.

── more in #ai-tools 4 stories · sorted by recency
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/show-hn-mira-search-…] indexed:0 read:1min 2026-05-28 ·