User repository on GitHub
Get FAR MORE done with FAR FEWER tokens 🔥
- Forks
- 44
- Last pushed
- Updated Oct 2026
Repo stats from the GitHub API, cached Oct 2026.
Jev directory
CommunityJev Tools & CLIs
Distill
A Rust coding-agent harness and TUI that uses Jev to pick a model and effort, route bounded utility work, and decide what context to keep, with code constraining every choice and validating results.
- Stars
- 723
- Language
- Rust
- License
- APACHE-2.0
- Reading time
- 3 min
- Published by
- Community
- Author
- samuelfaj
- Last verified
- Sep 2026
About this project
Distill is a lightweight coding agent harness and TUI, built to get far more done with far fewer tokens. It works with Grok and Codex subscriptions, and with any model that speaks the OpenAI-compatible protocol. Same quality and 9/9 passing runs. For an installation made with the release installer: Open Model tiers on the home screen, click change beside the current model, or enter /tiers. You can edit each tier in that screen.
Quoted from the project's README. Read the full README.
Works forJev for Routing
Used inJev for Software & DevTools
Highlights
- Three model tiers - reasoning, worker, and utility - each have a model and effort picker, and auto is the default.
- Jev selects the model and effort per turn and routes bounded utility tasks, while code constrains the choices and validates the results.
- Failed or low-confidence Jev decisions leave the normal path in place, and Jev never decides tool permissions.
- Works with Grok and Codex subscriptions and any OpenAI-compatible model, installed with a one-line curl or PowerShell script.
Quickstart
bash
curl -fsSL https://raw.githubusercontent.com/samuelfaj/distill/main/install.sh | sh
export PATH="$HOME/.local/share/distill/bin:$PATH"
distill --versionWatch out
Apache-2.0-licensed. Routing can send your request and recent steps to the configured endpoint, so read docs/jev-routing.md before enabling it.
Questions about this entry
Short answers drawn from the linked project or page.
No to both. Jev selects the model and effort per turn and routes bounded utility tasks, but code constrains the choices and validates the results; a failed or low-confidence decision leaves the normal path in place, and Jev never decides tool permissions. You pick the strongest model as main and a cheaper one as worker, and the main model delegates as much as it can do well.
Two rule sets are compiled into every system prompt, for the main model and its subagents, and neither has an off switch. Caveman makes the agent write terse prose while keeping code, paths and error strings exact, adjustable with /caveman lite, full or ultra, where a stored off is ignored. Ponytail makes it question whether the code needs to exist and reuse the codebase, standard library and installed dependencies first.
It is the author's own single run, not an independent evaluation: Distill at 308.18 seconds and about $0.19 estimated cost against Codex at 465.27 seconds and $0.59, with 9 of 9 passing runs on both sides. The settings differ too, with auto effort for two tiers against a fixed high effort for Codex.
Routing can send your request and recent steps to the configured endpoint, which is why the entry points at docs/jev-routing.md before you switch it on. Otherwise Distill works with Grok and Codex subscriptions and any OpenAI-compatible model, installs with a one-line curl or PowerShell script, and upgrades through `distill update`.
On Mac or Linux run the install.sh script with curl and add $HOME/.local/share/distill/bin to PATH; on Windows use the install.ps1 script and set the user Path. For an installation made with the release installer, distill update upgrades it.
The main model is required and owns every session, the optional worker model runs delegated assignments the main model hands it as specs, and the utility model handles bounded tasks like extraction, summaries, and compression. Each tier takes an effort level or auto, which is the default, set with /model, /worker-model, and /utility-model.
Apache-2.0. The README points at docs/jev-routing.md for how routing works before you enable it, docs/token-saver.md for where the savings come from, and docs/local-models.md for accounts and local models.
Reactions & coverage
Posts, threads, and videos about this entry from around the web.
X: A usage router for Grok Bot
X: A prompt box that fills itself in
X: How Jev makes agents faster and cheaper
YouTube: CJ from Syntax demos browser use, classification, code review, a model router, and a chat bot with no LLM.
More Jev resources like this
Rust coding-agent harness (MIT, ~20k stars) whose crates/jcode-base/src/jev.rs is a typed TypeSafe Jev decision transport, used for memory pruning (memory_jev.rs), browser navigation (browser_jev.rs), and voice routing (voice_intent.rs) with separate provider selectors per purpose.
Jev Tools & CLIs#community#rust#coding-agent
Community
1jehuang/jcodeRust
1jehuang/jcodeRustA local-first Herdr CLI that filters your agent subscriptions with fixed eligibility rules, then asks TypeSafe Jev to rank the remaining Cursor, Claude Code, Codex, or OpenCode options and pick a reasoning effort.
Jev Tools & CLIs#community#typescript#cli
Community
nidhi-singh02/agent-routerTypeScript
nidhi-singh02/agent-routerTypeScriptA TypeScript layer beside an LLM: JevSift selects approaches, JevLoop audits progress with typed actions, and JevLong watches sessions for stalls without steering them.
Jev Tools & CLIs#community#typescript#coding-agent
Community
Alex314618-create/JevRevTypeScript
Alex314618-create/JevRevTypeScriptA router that turns models, subagents, skills, MCP tools, CLIs, and plugins into one candidate set Jev answers with a single typed Choice, while the router code owns availability, permissions, and confirmation.
Jev Tools & CLIs#community#typescript#routing
Community
BillionsBobby/JevRouterTypeScript
BillionsBobby/JevRouterTypeScriptA CLI that scores code quality by asking Jev yes/no questions about each file, runs your existing test command, and turns uncovered paths into small actionable queries for agents.
Jev Tools & CLIs#community#rust#cli
Community
supercorp-ai/supercovRust
supercorp-ai/supercovRustA Rust CLI that ranks agent skills for the next step with a two-stage Jev pass plus lexical prefiltering, with Claude Code hooks, a TUI, abstention, and --why-not explanations.
Jev Tools & CLIs#community#rust#cli
Community
Dicklesworthstone/skillrankerRust
Dicklesworthstone/skillrankerRustFrom 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.