User repository on GitHub
Fish-style zsh history autosuggestions, ranked by an AI model instead of prefix matching
- Forks
- 1
- Last pushed
- Updated Oct 2026
Repo stats from the GitHub API, cached Oct 2026.Homepage
Jev directory
CommunityJev Tools & CLIs
Guesswork
Fish-style zsh autosuggestions ranked by Jev instead of prefix matching: as you type, it asks a model which recent command you are most likely retyping and shows the best guess as grey inline text you accept with the right arrow.
- Stars
- 3
- Language
- TypeScript
- License
- MIT
- Reading time
- 3 min
- Published by
- Community
- Author
- findmalek
- Last verified
- Oct 2026
Works forJev for Ranking
Used inJev for Software & DevTools
- Languages
- TypeScript
- Platforms
- CLICloudflare
- Capabilities
- Open sourceRequires API key
Highlights
- Install with curl -fsSL https://guesswork.findmalek.com/install.sh | bash; the README badges name Node 22+ and zsh 5.9+.
- Four providers: TypeSafe or Cloudflare Workers AI run the real Jev model; Anthropic or Groq work as fallbacks.
- Understands abbreviations and fuzzy matches - gst resolves to git status, and no entry needs to start with your input.
- Costs about $0.000081 per Jev request; the README estimates a few hundred requests a day for a year at around $10.
- Hooks zle directly with no raw escape codes or terminal-specific code paths, so it runs anywhere zsh does.
Quickstart
bash
curl -fsSL https://guesswork.findmalek.com/install.sh | bashWatch out
Grounded in the repository README and GitHub metadata reviewed on 2026-10-15; the per-request cost and the install badge (about 1,674 clones, which the README calls a close proxy) are the project's own figures, no accuracy benchmark is published, and the README's provider section was truncated mid-sentence.
Questions about this entry
Short answers drawn from the linked project or page.
Prefix matching only works if you remember how a command started, and fuzzy recollection like something with --dir blog last week defeats it. guesswork asks an AI model which recent command you are most likely retyping, covering abbreviations and fuzzy matches, and shows the best guess inline.
Four: TypeSafe or Cloudflare Workers AI run the real Jev model, and Anthropic or Groq work as fallbacks if you would rather not sign up for either. You configure a provider and key, and suggestions are ranked by the model rather than by prefix.
Run curl -fsSL https://guesswork.findmalek.com/install.sh | bash. There is no build step - it is plain TypeScript run directly by Node - and the README badges list Node 22+ and zsh 5.9+ as requirements.
The README prices Jev at about $0.000081 per request and estimates a few hundred requests a day for a year at around $10 total. Fallback providers carry their own pricing, which the README does not restate, and any provider needs its key configured.
Suggestions appear fish-style as grey text after your cursor and you press the right arrow to accept them, working in both emacs and vi keymaps. The README's examples show gst resolving to git status and last 5 commits resolving to git log --oneline -5, each with a displayed score.
No. The demo recording is real, not a mockup, but the regeneration scripts run against a fabricated history, never your real one, and need vhs, ffmpeg, and a provider key. The GIF in the README was recorded from that same fabricated environment.
It is findmalek's open-source MIT project, listed on madewithjev as Malek Gara-Hellal's work. The README publishes no accuracy benchmarks, the clone badge is explicitly a close proxy rather than a byte-exact install count, and suggestions depend on a reachable provider.
More Jev resources like this
Write end-to-end tests for web apps in plain English; jev-e2e turns each case into a plan, has Jev pick the page controls, and runs it with Playwright, returning PASS, FAIL or BLOCKED with HTML, JSON and masked screenshot evidence.
Jev Tools & CLIs#community#catalog-import#madewithjev
Community
perixtar/jev-e2eTypeScript
perixtar/jev-e2eTypeScriptCLI and agent skill that finds useful notes in a local Obsidian vault or Logseq Markdown graph: it shortlists candidates on your machine, then has Jev rank them against your question and returns the exact source passage.
Jev Tools & CLIs#community#catalog-import#madewithjev
Community
emlembow/jev-graph-searchJavaScript
emlembow/jev-graph-searchJavaScriptA Chrome extension that reorders the Gmail message list on screen only - unread first, critical messages on top - using Jev to score each row's urgency and criticality and to answer one boolean question per custom label.
Jev Tools & CLIs#community#catalog-import#madewithjev
Community
iamomiid/jev-inboxTypeScript
iamomiid/jev-inboxTypeScriptA bash-native CLI from Parall that gives an agent Jev as an atomic tool: pipe any event stream through a plain-English question and it prints only the lines that pass, composing with the tools your agent already pipes.
Jev Tools & CLIs#community#catalog-import#madewithjev
Community
parall-hq/jevableTypeScript
parall-hq/jevableTypeScriptA toolkit that turns an agent skill into a project-specific Oxlint plugin: deterministic AST checks handle structural rules while Jev answers the narrow typed questions that need contextual judgment.
Jev Tools & CLIs#community#catalog-import#madewithjev
Community
cephalization/jev-oxlintTypeScript
cephalization/jev-oxlintTypeScriptExperimental deterministic model routing for coding agents with Jev in the loop: assign capabilities to models and copy one workflow prompt into Codex Desktop. The author's own Pokédex pilots did not show a cost win and he recommends against adopting it.
Jev Tools & CLIs#community#catalog-import#madewithjev
Community
instructa/switchloomTypeScript
instructa/switchloomTypeScriptFrom the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.