What is the WrongStack coding agent?

A free, open-source AI coding agent written in TypeScript from scratch — its own kernel, provider transports with SSE, tool executor, permission policy, memory system and multi-agent runtime, rather than a wrapper around another CLI.

How do I install WrongStack?

On macOS or Linux run curl -fsSL https://wrongstack.com/install.sh | sh; on Windows use the PowerShell one-liner irm https://wrongstack.com/install.ps1 | iex. It ships as one self-contained binary and does not need Node.js.

What are the Brain, SAGE, Kanban, mailbox and Chimera?

The Brain evaluates policy decisions, SAGE retains project knowledge across sessions, Kanban tracks work and verification, the mailbox links agents across clients, and Chimera reviews agent diffs.

Which models can WrongStack use?

Any provider via API keys or subscription sign-in, a live models.dev catalog, and local options such as Ollama, vLLM and LM Studio presets; per-role routing with fallback chains is supported.

Can WrongStack run with Claude Code?

Yes. Per the release notes, an installed ACP agent can be the session leader through wstack --provider acp --model claude-code, and the delegate tool can hand tasks to a Claude Code worker; the older anthropic-oauth login was removed.

Is my data sent anywhere?

The README says memory, tools, providers, permissions and the multi-agent runtime all work on your machine with no upstream agent to phone home to. Secrets are encrypted and a permission policy gates every tool call.

Does WrongStack cost anything?

The agent itself is free and MIT licensed with no subscription required. You pay only for whatever model access you choose — your existing API keys, a subscription sign-in, or nothing if you run a local model.

Jev directory

CommunityJev Tools & CLIs

WrongStack

A free, MIT-licensed AI coding agent that reads code, edits files, runs commands and reasons through bugs across a terminal REPL, a full-screen TUI and a browser UI, with visible permission boundaries and durable project memory.

See highlights Open github.com

Stars: 380
Language: TypeScript
License: MIT
Reading time: 2 min
Published by: Community
Author: wrongstack
Last verified: Oct 2026

wrongstack/wrongstack

Organization repository on GitHub

View on GitHub

An AI coding agent that reads your code, edits files, runs commands, and reasons through bugs — across a terminal REPL, a full-screen TUI, and a browser UI, while you keep your hand on every permission.

Forks: 40
Last pushed: Updated Oct 2026

Repo stats from the GitHub API, cached Oct 2026.Homepage

Tagscommunity catalog-import logicrw coding-agent memory

Works forJev for Code Jev for Tool calling

Used inJev for Software & DevTools

Languages: TypeScript
Platforms: CLI SQLite
Capabilities: Open sourceSelf-hostedRuns locally

Highlights

Installs as one self-contained binary for macOS and Linux, with a PowerShell one-liner for Windows; no Node.js needed.
SAGE keeps long-term project memory in SQLite/FTS5 anchored to files, symbols, commands and commits, re-verified as they change.
Six surfaces — REPL, TUI, WebUI, SimpleUI, Desktop and the cross-machine HQ — share the same sessions and memory.
The Brain rules on risky policy calls, Kanban tracks work and verification, and Chimera reviews agent diffs.
Model-agnostic: a models.dev catalog, Ollama / vLLM / LM Studio presets, API keys or subscription sign-in, and per-role routing with fallback chains.

Quickstart

bash

curl -fsSL https://wrongstack.com/install.sh | sh

Watch out

Copy reviewed against the repo README and GitHub metadata on 2026-10-15; release notes are the project's own and the README states no benchmarks. The README notes the old anthropic-oauth login was removed in favour of an Anthropic API key or ACP.

Questions about this entry

Short answers drawn from the linked project or page.

Reactions & coverage

Posts, threads, and videos about this entry from around the web.

Hacker News: jevmem keeps Claude Code project memory by asking Jev which turns are worth saving.

avinashjetwani61 pointsJevmem – automatic project memory for Claude Code, built on JevRead the thread on Hacker News

X: Decision outputs in Grep

X: A usage router for Grok Bot

X: The compaction plugin, in Chinese

From the Jev guides

Original write-ups on what this entry is for.

Jev at home: voice pipelines, Assist routing, and sentinels

Oct 16, 2026

How Home Assistant conversation agents, macOS voice pipelines, and an advisory sentinel put a fast typed decision in the loop — one call per utterance, thresholds in code, and a daily token budget.

Jev Tools & CLIs#home-assistant#voice#automation

Read article

Search visibility, audited by decisions

Oct 16, 2026

Jev-based SEO tools split the crawl from the call: deterministic crawlers count and parse, while typed judgments rank fixes, place internal links, score search results, and track AI-answer mentions — with costs and caveats attached.

Jev Tools & CLIs#seo#search-visibility#audits

Read article

Jev on the Gateways: Vercel, Workers AI, and OpenRouter

Oct 12, 2026

The request shape barely changes across surfaces; the surrounding contract does. When to call the API directly and when a gateway, a Worker binding, or a router entry earns its extra hop.

Jev Tools & CLIs#gateway#vercel#cloudflare

Read article

From the Jev AI model community

Posts from builders shipping with Jev right now.

Follow @typesafeai

WrongStack

Highlights

Quickstart

Questions about this entry

What is the WrongStack coding agent?

How do I install WrongStack?

What are the Brain, SAGE, Kanban, mailbox and Chimera?

Which models can WrongStack use?

Can WrongStack run with Claude Code?

Is my data sent anywhere?

Does WrongStack cost anything?

Reactions & coverage

Related terms

More Jev resources like this

From the Jev guides

From the Jev AI model community