afwf_md5 0.1.2

Alfred Workflow to generate random md5, sha256, sha512 hash and calculate file hashes.

---

afwf_md5 — MD5 / SHA-256 / SHA-512 Alfred Workflow

An Alfred Workflow that computes MD5, SHA-256, and SHA-512 checksums of local files directly from Alfred. No terminal needed — type a path, get a hash, copy it with ⌘C.

When the query is empty, the workflow generates 10 random hashes so you can use it as a secure random string generator.

Install

1. Make sure you have Alfred 5+ installed with the Power Pack.

2. Go to Releases and download the latest .alfredworkflow file.

3. Double-click the file to install.

Usage

Compute MD5

Type md5 followed by an absolute file path:

md5 /path/to/file.zip

• File found — returns the MD5 hex digest. Press ⌘C to copy.

• Empty query — generates 10 random MD5 hashes (useful as secure random strings).

• Directory — shows an error item (directories are not supported).

• Path not found — shows an error item.

Compute SHA-256

Type sha256 followed by a path:

sha256 /path/to/file.zip

Compute SHA-512

Type sha512 followed by a path:

sha512 /path/to/file.zip

CLI Reference

The workflow is backed by a Python CLI (afwf-md5) with one subcommand per algorithm. Alfred’s Script Filter calls these directly.

Subcommand	Alfred keyword	Output
afwf-md5 md5 --query '{query}'	md5	MD5 hex digest (32 chars)
afwf-md5 sha256 --query '{query}'	sha256	SHA-256 hex digest (64 chars)
afwf-md5 sha512 --query '{query}'	sha512	SHA-512 hex digest (128 chars)

Dev Script field (local .venv):

.venv/bin/afwf-md5 md5 --query '{query}'

Production Script field (via uvx, no virtualenv needed):

~/.local/bin/uvx --from afwf_md5==0.1.1 afwf-md5 md5 --query '{query}'

Error Log

Runtime errors are written to:

~/.alfred-afwf/afwf_md5/error.log

If a Script Filter returns an error item, pressing Enter opens the log file directly.

Development

# Install dependencies
mise run inst

# Run tests with coverage
mise run cov

The core logic lives in two files:

• afwf_md5/hashes.py — pure function main(query, hash_algo) → ScriptFilter, no Alfred dependency, fully unit-testable.

• afwf_md5/cli.py — Command class wired to fire.Fire; each method calls hashes.main() and sends feedback to Alfred.