Zero

23 | Content Creator & Dev / Ex Microsoft

[Zero @zer0point_eth]

HOLY SH*T! 10 Jev repos that turn Opus 5.5 into a cheaper, safer coding setup

Opus 5.5 builds. but you don’t want it burning tokens on every small decision around the build. these 10 repos hand those decisions to Jev, a fast, cheap model that only answers bounded questions: which model should take this task, did the work really get finished, does this edit break a team rule

01 model-router → hooks into subagent spawning in Claude Code and asks Jev how hard the task is: lookups go to Haiku, bounded judgment to Sonnet, open-ended work to Opus. the default mode only logs its pick, auto mode actually sets the model
https://t.co/rCBcBztO1w

02 effortless → a bar above the Claude Code prompt that sets reasoning effort for every message and shows context fill plus a prompt-cache countdown, with one-click handoff or compact. Haiku is the default judge, Jev is an optional faster one that Haiku backs up when it’s unsure
https://t.co/ZY7Nyp9Lbu

03 jev-lint → after every edit by Claude Code or Codex, a hook checks the new lines against your team’s rules in about 0.3 s, things like tests that can never fail or user IDs trusted from the client, so the agent fixes them while the file is still open
https://t.co/n9legbigQw

04 Jevris → a local control layer that sits next to Claude Code, Codex, Kilo, OpenCode or Antigravity. facts, permissions and verification stay in plain code, Jev only answers bounded questions, and a task counts as done only when independent evidence says so. release candidate, install from source for now
https://t.co/ZJeIM5SY5d

05 visual-proof → for Vite apps: a watcher re-shoots the screens you touched on every save, and finish packs the stills (plus a replay video if ffmpeg is installed) into a proof block, failing if a changed screen has no clean frame. optional model checks compare the stills with the claim you wrote, and Jev prunes routes and writes captions
https://t.co/TT2MZXVKjm

06 CloakBrowser Agent → give it a goal in plain language: Jev picks each click or keystroke in about 0.3 s, a stealth Chromium performs it with human-like input, and the coding agent gets back only the relevant page sections as markdown. MCP server, CLI and Python API, CAPTCHAs not handled
https://t.co/E2LaNAdPMr

07 jev-doc-search → no vector DB: PageIndex turns a long PDF into a tree of sections, and Jev picks one branch per level until it lands on the pages that answer the question. tested on a 93-page and a 318-page annual report
https://t.co/UOKwx7491S

08 ex-regex → a Python library shaped like re, except the match test is meaning: sub, findall and extract take a plain-English description. a finder proposes spans, Jev scores them, and every result is a verbatim slice of your input with offsets
https://t.co/ud63pNPg7j

09 anofox-decide → a DuckDB extension that adds decide_probability, decide_choice and decide_table to SQL, so you can score or classify rows with Jev. remote calls stay off until you opt in, and local models can run fully offline
https://t.co/lifYmZLmqx

10 jevals → agent evals and guardrails from a single request: the quickstart packs six checks (tool choice, use of tool results, grounding, scope, indirect injection, PHI) into one call, and the same checks can act as gates that allow, escalate or block a tool call. alpha
https://t.co/uuBPjjhu4D

the one repo here with a clean before/after is jev-lint. its author ran the same tasks with the hook off and on:

Claude Code, 96 runs → 2.38 rule violations per task without the hook, 1.21 with it
Codex, 48 runs → 1.67 without, 0.92 with it

that’s roughly half the violations, for about $0.04 extra per task. a violation caught at the edit costs a few seconds. the same one caught in review costs a whole fix and re-review round

that’s the job for the cheap fast model: check the edit while the agent still has the file open

keep Opus 5.5 for building and hard reasoning. let Jev answer small, bounded questions, and let plain code enforce the result and handle anything the model isn’t sure about

pick the repo that removes your biggest bottleneck

[Zero @zer0point_eth]

@me_cool_off you're welcome bro

[Zero @zer0point_eth]

HOLY SH*T! 10 Jev repos that turn Opus 5.5 into a cheaper, safer coding setup

Opus 5.5 builds. but you don’t want it burning tokens on every small decision around the build. these 10 repos hand those decisions to Jev, a fast, cheap model that only answers bounded questions: which model should take this task, did the work really get finished, does this edit break a team rule

01 model-router → hooks into subagent spawning in Claude Code and asks Jev how hard the task is: lookups go to Haiku, bounded judgment to Sonnet, open-ended work to Opus. the default mode only logs its pick, auto mode actually sets the model
https://t.co/rCBcBztO1w

02 effortless → a bar above the Claude Code prompt that sets reasoning effort for every message and shows context fill plus a prompt-cache countdown, with one-click handoff or compact. Haiku is the default judge, Jev is an optional faster one that Haiku backs up when it’s unsure
https://t.co/ZY7Nyp9Lbu

03 jev-lint → after every edit by Claude Code or Codex, a hook checks the new lines against your team’s rules in about 0.3 s, things like tests that can never fail or user IDs trusted from the client, so the agent fixes them while the file is still open
https://t.co/n9legbigQw

04 Jevris → a local control layer that sits next to Claude Code, Codex, Kilo, OpenCode or Antigravity. facts, permissions and verification stay in plain code, Jev only answers bounded questions, and a task counts as done only when independent evidence says so. release candidate, install from source for now
https://t.co/ZJeIM5SY5d

05 visual-proof → for Vite apps: a watcher re-shoots the screens you touched on every save, and finish packs the stills (plus a replay video if ffmpeg is installed) into a proof block, failing if a changed screen has no clean frame. optional model checks compare the stills with the claim you wrote, and Jev prunes routes and writes captions
https://t.co/TT2MZXVKjm

06 CloakBrowser Agent → give it a goal in plain language: Jev picks each click or keystroke in about 0.3 s, a stealth Chromium performs it with human-like input, and the coding agent gets back only the relevant page sections as markdown. MCP server, CLI and Python API, CAPTCHAs not handled
https://t.co/E2LaNAdPMr

07 jev-doc-search → no vector DB: PageIndex turns a long PDF into a tree of sections, and Jev picks one branch per level until it lands on the pages that answer the question. tested on a 93-page and a 318-page annual report
https://t.co/UOKwx7491S

08 ex-regex → a Python library shaped like re, except the match test is meaning: sub, findall and extract take a plain-English description. a finder proposes spans, Jev scores them, and every result is a verbatim slice of your input with offsets
https://t.co/ud63pNPg7j

09 anofox-decide → a DuckDB extension that adds decide_probability, decide_choice and decide_table to SQL, so you can score or classify rows with Jev. remote calls stay off until you opt in, and local models can run fully offline
https://t.co/lifYmZLmqx

10 jevals → agent evals and guardrails from a single request: the quickstart packs six checks (tool choice, use of tool results, grounding, scope, indirect injection, PHI) into one call, and the same checks can act as gates that allow, escalate or block a tool call. alpha
https://t.co/uuBPjjhu4D

the one repo here with a clean before/after is jev-lint. its author ran the same tasks with the hook off and on:

Claude Code, 96 runs → 2.38 rule violations per task without the hook, 1.21 with it
Codex, 48 runs → 1.67 without, 0.92 with it

that’s roughly half the violations, for about $0.04 extra per task. a violation caught at the edit costs a few seconds. the same one caught in review costs a whole fix and re-review round

that’s the job for the cheap fast model: check the edit while the agent still has the file open

keep Opus 5.5 for building and hard reasoning. let Jev answer small, bounded questions, and let plain code enforce the result and handle anything the model isn’t sure about

pick the repo that removes your biggest bottleneck

[Zero @zer0point_eth]

Jev + Claude... I still can't believe how broken this combination is

I packed the reasoning of Claude and the speed of Jev into ONE workflow - now 99% of traditional AI coding agents look slow, dumb, and expensive

The pipeline is completely seamless:

prompt → Claude plans → Jev decides & scopes → Claude writes diff → Jev validates → result

→ Jev strips routing from Claude: the LLM never decides *which* file to touch or *when* to retry; it only writes the implementation

→ Pre-flight context pruning: instead of dumping an entire repository into Claude's context, Jev isolates only the 2-3 required files in ~35 ms

→ Bounded terminal safety: shell commands, npm installs, and git operations pass through a typed Jev gate before executing

→ Zero-hallucination verification: a Jev Noul check mathematically confirms that every function and import in the diff exists in your codebase

→ Hard confidence gates: ≥0.90 confidence merges automatically, 0.60-0.89 requests a 1-click review, <0.60 halts immediately

→ Runs directly inside Claude Code or Cursor while cutting token consumption by ~75%

Most developers give Claude raw terminal access and pray it doesn't break production

Pairing it with Jev turns an unpredictable chatbot into a disciplined software engineer everyone will call "obvious" by 2030

[Zero @zer0point_eth]

@kinderlayer appreciate it

[Zero @zer0point_eth]

@0xKorens gotta stay ahead of the game at all times

[Zero @zer0point_eth]

@Lummox_eth this is pure gold!

[Zero @zer0point_eth]

@0xKorens anyone could be in his shoes. that’s great!

[Zero @zer0point_eth]

@paonx_eth is that the whole prompt or is there more?

[Zero @zer0point_eth]

@me_cool_off @variational_io @JustinCBram more competition means a higher price per point too, right?

[Zero @zer0point_eth]

@starmexxx open sourced?

[Zero @zer0point_eth]

@morpphhhaw this is literally the perfect employee. never eats, never sleeps, always does what i want lol

[Zero @zer0point_eth]

@0xRicker so what can’t the grok bot do?

[Zero @zer0point_eth]

@kinderlayer tbh I like both options

[Zero @zer0point_eth]

@0xShoopy wonder if they get paid extra for making these guides

[Zero @zer0point_eth]

@zodchiii the first thing i do whenever i open a new chat

[Zero @zer0point_eth]

@distortgeekin bro, you’re literally a genius!

[Zero @zer0point_eth]

@scrblanc bro, this is literally what i’ve been searching for for so long. really appreciate it

[Zero @zer0point_eth]

@Lummox_eth already tested 6 repos, this stuff is wild

[Zero @zer0point_eth]

@tekari0 wait, are you the student?

[Zero @zer0point_eth]

@goodworse pretty sure openai has already figured out how to outdo them