2b-agent 2.1.2

A local-first coding agent that keeps small local models focused instead of hallucinating.

2B

A local-first coding agent for the terminal. It runs your own models over Ollama's native protocol, keeps them focused instead of hallucinating, and gives you a full-screen TUI — streaming replies, a live plan checklist, narrated tool actions — without ever routing your local model through a translation layer that would confuse it.

I built 2B to keep working when the power and the internet don't — I live somewhere the grid isn't a guarantee, and I wanted a coding agent that doesn't fall apart the moment I'm offline.

macOS only. 2B is built and tested for macOS — the installer is a shell script that leans on Homebrew, and the clipboard integration uses pbcopy. It hasn't been tested elsewhere.

---

Why I built it

I kept being told that small models — Nemotron 3 Nano 4B, gpt-oss:20b, the Qwen family — were "good enough" for agentic coding. On paper they were. In practice, every off-the-shelf harness I tried broke them:

• opencode made gpt-oss invent tool names that didn't exist and emit fake <command> tags as plain text.
• Cline, Goose, Continue, OpenHands each failed in their own way — malformed tool schemas, reasoning collapse, the model narrating tool calls instead of making them.

The models weren't the problem. The harnesses were. Nearly all of them talk to a local model through a generic OpenAI-compatible /v1 shim and pile abstraction on top of it. That shim measurably degrades a small model's tool selection, and the extra complexity buries whatever capability the model actually has.

The one thing that worked cleanly was a ~350-line script I wrote that talked to Ollama's native /api/chat endpoint with a tiny, fixed set of five tools. So I grew that prototype into a real, shareable tool. That's 2B.

The core rule, and the whole point: all complexity lives on the host side. The model's world never changes — the same five tools, the same native wire format for whatever provider is active, no generic shim. Everything you see below — the TUI, the plan checklist, task management, model switching, auto-compaction — is something 2B renders around the tool loop, never a new thing the model has to understand.

---

What it does

The one idea that makes it work: the model only ever sees five tools — list_files, read_file, search_files, edit_file, write_file — over its provider's native wire format, never a generic /v1 shim. Every feature below is something 2B renders around that tiny loop, so a small model's world stays dead simple while you get a full-screen coding agent.

Reliable with small models — the host does the hard part:

• Edits that survive drift. edit_file matches in tiers (exact → whitespace-tolerant → indentation-agnostic), so a near-miss still lands — never on an ambiguous match. The tool the model sees is unchanged; the tolerance is all host-side.
• Catches its own mistakes. After an edit, 2B runs the file's checker (dart analyze, ruff, …) and folds new errors into the same tool result — the model sees the break it just caused, no new tool to learn.
• Finds definitions, not just matches. search_files floats a symbol's definition to the top and read_file appends a symbol outline — resolved semantically over a language server (LSP) when one's installed, with a dependency-free regex fallback otherwise.
• Rescues weak tool-calls. Recovers calls a small model emits as plain text, and nudges one that narrates a plan but forgets to act — so weak models actually finish the job.

A full-screen experience, not a log dump:

• Streaming TUI with a live plan checklist (□ ■ ✓) and a status line showing a context meter (ctx N%) plus, for local models, a live RAM/GPU readout.
• Narrated actions — plain-language steps with a ✓/✗ tree, not raw read_file {...}:

  ├ ✓ Searching for "MemoryScopeLevel" in lib
  └ ✓ Editing lib/memory/memory_store.dart
  

• Themes (/theme), copy that works (Ctrl+Y, or drag + Ctrl+C via pbcopy), keyboard scroll (Shift+↑/↓), and queued/background tasks (Ctrl+B, /fg).

Any model, one conversation. Ollama (local + cloud), OpenAI, OpenRouter, Mistral, NVIDIA, DeepSeek, Cerebras, Anthropic, and Google Gemini — each over its own native format, all streaming. History is provider-agnostic, so you switch models mid-task with /model and keep every bit of context: start on a local Qwen, hand it to Claude when it gets hard.

Keeps the thread — or doesn't, your call. Cloud sessions carry context across messages by default; local models stay lean until you opt in with /continuity. /new starts a fresh thread, and /export dumps the whole session — tool calls included — to a Markdown file.

Knows your project. /init writes a compact 2B.md (stack, layout, ranked symbol map) that's auto-loaded into context; /map shows an outline on demand — all bounded so it never floods a small window.

Runs things, safely. Local models get run_git (git only — no raw shell); cloud models get a full run_command whose writes are sandbox-confined to the workspace (macOS sandbox-exec / Linux bubblewrap). Network and secret-path commands re-prompt even under "allow-all," your ambient secrets (*_API_KEY, AWS_*) are scrubbed from the child env, and tool output is fenced as untrusted so a poisoned file can't hijack the model.

Scales without bloating the model. Cloud models can delegate read-only investigations (and isolated edit sub-tasks) in parallel so a big search never bloats the main thread; MCP tools are opt-in per tool; auto-compaction folds old turns into a summary near the window limit; and sessions persist to SQLite (2b --continue / --resume, /sessions). /undo reverts the last write or edit.

Three modes (Shift+Tab): normal (confirm writes) · accept edits (auto-apply) · plan (read-only — investigate and return a plan, touching nothing).

---

Install

One line — paste it in your terminal:

curl -fsSL https://raw.githubusercontent.com/dea6cat/2b-agent/main/install.sh | sh

It installs uv if you don't have it, installs the 2b command, then — on an interactive terminal — walks you through local model setup:

1. Optional clean install — offers to remove other agentic tools that proved unreliable with local models (opencode, Continue, Goose, Cline, OpenHands) and their configs. Off by default; it asks first.
2. Grades your machine — reads your RAM and chip and rates each candidate model (✓ fits well / ~ tight / ✗ needs NGB+), defaulting to the best one your hardware can run.
3. You pick one or several from the menu.
4. Installs Ollama and pulls what you chose, with a live progress bar.
5. Self-tests each model — tok/s + GPU residency, then a correctness check that runs a real one-line edit through 2B itself and verifies the result (✓ correct / ✗ wrong, ~20–90s per model). It only reports — it never removes a model — and --no-benchmark skips it. Then it prints how to launch 2B.

Already have Ollama and some models? It skips what you already have — it lists your installed models, offers to just use them (pulling nothing), and marks anything in the menu you've already got. Your existing setup is left untouched.

Prefer to do it by hand? Install the published package from PyPI, then run the same onboarding the installer uses:

pip install 2b-agent      # or: uv tool install 2b-agent
2b setup                  # grades your machine, installs Ollama, pulls a model, self-tests, fixes PATH

On macOS/Linux you can also use Homebrew (it puts 2b on your PATH automatically):

brew install dea6cat/2b/twob-agent   # the formula is twob-agent; it installs the `2b` command
2b setup

2b setup is the single source of truth for onboarding — the curl … | sh installer just installs uv + the 2b command and then runs it, so you get the exact same setup either way. (On first launch with no model configured, 2b offers to run it for you.)

The installer — and 2b setup — are scriptable: --yes (accept defaults, no prompts), --clean / --no-clean, --models "qwen3.5:9b qwen3:8b", --no-models, --no-benchmark (skip the correctness check), --fix-path / --no-fix-path (add uv's tool dir to your PATH for you via uv tool update-shell, or leave it — otherwise it asks, and never edits a profile without consent). Pass them through the pipe with ... | sh -s -- --yes --models "qwen3.5:9b".

---

Use it

2b                       # start in the current directory, autodetects a local model
2b "add a docstring to lib/main.dart"   # run one task, then drop into the session
2b --model qwen3.5:9b    # pin a model
2b --continue            # resume your most recent session (--resume <id> for a specific one)
2b --list-sessions       # list saved sessions
2b --list-models         # what's available across configured providers
2b --doctor              # diagnose PATH, Ollama, and the default model, then exit
2b --test                # grade installed models (tok/s + a real coding test) and compare
                         #   them to the latest coding models on ollama.com
2b --test auto           # also remove failing models, then pull + coding-test the best new
                         #   candidate and recommend it only if it passes
2b --update              # upgrade to the latest release (uv tool upgrade)
2b --rm                  # uninstall 2B and delete its config (asks first); --rm --yes to skip

Then just type what you want done. Type / to see the commands.

Updating

One command, whatever you installed with — it detects the method and runs the right upgrade:

2b --update

That resolves to uv tool upgrade 2b-agent (the curl … | sh installer / uv), pipx upgrade 2b-agent (pipx), or pip install -U 2b-agent (pip). You can of course run the matching command yourself — e.g. if you installed with pip:

pip install -U 2b-agent

2B also checks for a newer release in the background (at most once a day, never blocking startup) and prints a one-line notice on the next launch when one is available — set TWOB_NO_UPDATE_CHECK=1 to turn that off. Releases are tagged vMAJOR.MINOR.PATCH.

Providers

Local Ollama needs nothing. For anything else, set the matching environment variable and it shows up automatically in /models:

Provider	Environment variable
Ollama	OLLAMA_API_BASE (or OLLAMA_HOST)
Ollama Cloud	OLLAMA_API_KEY
OpenAI	OPENAI_API_KEY
OpenRouter	OPENROUTER_API_KEY
Mistral	MISTRAL_API_KEY
NVIDIA	NVIDIA_API_KEY
DeepSeek	DEEPSEEK_API_KEY
Cerebras	CEREBRAS_API_KEY
Anthropic	ANTHROPIC_API_KEY
Google	GEMINI_API_KEY (or GOOGLE_API_KEY)

Or connect one from inside 2B — /connect <provider> prompts for the key with a hidden field and saves it to ~/.config/2b/keys.json (chmod 600) so it's there next time; /connect on its own shows what's connected, and /disconnect <provider> removes a saved key. A key exported in your shell always takes precedence over a saved one.

Switch models anytime with /model <name>. A bare name works when it's unambiguous; otherwise use provider:model (e.g. /model anthropic:claude-sonnet-5).

Commands

Command	What it does
/help	List commands
/model [name]	Show or switch model — context is preserved
/models [filter]	List available models, grouped by provider
/default [name]	Show or set the persisted default model (used when no --model is given)
/connect [provider] [key]	Connect a provider (hidden key prompt); bare shows status
/disconnect <provider>	Remove a saved provider key
/init	Scan the project → write 2B.md (a compact map auto-loaded into context on new tasks)
/map [subdir]	Show a budget-bounded symbol outline of the project
/mcp	MCP servers/tools: status, tools <server>, enable/disable <server> <tool…|all>
/mode [normal|accept|plan]	Set operating mode (or Shift+Tab to cycle)
/theme [system|light|dark]	Switch color theme
/context	Show estimated context usage (auto-compacts near the limit)
/continuity [on|off]	Carry conversation context across messages — on by default for cloud, opt-in for local
/new	Start a fresh conversation thread (keeps the scrollback on screen)
/export [path]	Export the whole session — tool calls and errors included — to a Markdown file
/copy	Copy the last reply to the clipboard (Ctrl+Y)
/task <desc>	Queue a task
/tasks	List tasks and their status
/fg <id>	Foreground a backgrounded task
/sessions	List saved sessions (resume with 2b --continue / --resume <id>)
/tool <name> <args>	Run a frozen tool directly, bypassing the model (e.g. /tool read_file path=a.dart)
/history search <q>	Search the scrollback; then n / N jump between matches
/yes	Toggle accept-edits mode
/undo	Revert the last write/edit
/diff	Re-show the last diff
/add <file>	Pre-load a file into the current task's context
/fetch <url>	Fetch a web page and pre-load its readable content into the current task's context
/clear	Reset the current task's history
/quit	Exit

Keyboard

Key	Action
Shift+Tab	Cycle operating mode
Shift+↑ / ↓	Scroll the conversation log (a line); PageUp / PageDown by a page
Ctrl+B	Background the running task
Ctrl+Y	Copy the last reply
Ctrl+C	Copy the current mouse selection
Esc	Stop the current stream/task immediately — back to idle
Ctrl+D	Quit
Tab	Accept the top /-command suggestion

MCP servers (extra tools)

2B can pull in tools from MCP servers (stdio) like dart or mempalace. But its whole reason for existing is that small local models break when you flood them with tools — so MCP tools are opt-in and curated per tool: you enable a server and pick exactly which of its tools reach the model. Nothing is exposed until you say so.

Declare servers the usual way — a Claude-Code-style mcpServers block in ~/.config/2b/mcp.json (or ./.mcp.json in a project, which wins):

{
  "mcpServers": {
    "dart": { "command": "dart", "args": ["mcp-server"] }
  }
}

Then curate from inside 2B:

/mcp                          # servers and how many tools each has enabled
/mcp tools dart               # list a server's tools ([x] = enabled)
/mcp enable dart hot_reload analyze_files
/mcp enable dart all          # expose everything (careful on small models)
/mcp disable dart hot_reload  # or /mcp disable dart  to turn the whole server off

Enabled tools appear to the model as server__tool (e.g. dart__hot_reload) and their results come straight back into the loop. Only the tools you enable are ever sent — the model's tool list stays as small as you keep it.

Configuration

• Context window (local) — sized to your machine. For a local model 2B works out the largest window your box can run comfortably and pins num_ctx to it on every request (Ollama otherwise defaults to ~4k regardless of the model). It reads the model's architecture and trained max from /api/show, computes the KV-cache cost per token, and fits it into the RAM left after the model weights plus a headroom reserve — so a 16 GB laptop, an 18 GB one, and a 64 GB workstation each get a different, appropriate window (e.g. qwen3.5:9b ≈ 13k on 18 GB), never more than the model was trained for. That number drives auto-compaction (~75%) and the read-a-section threshold. Set TWOB_CONTEXT_TOKENS to override (higher if you want to spend more RAM, lower to save it).

• Environment toggles. Everything on-by-default can be turned off: TWOB_CONTEXT_TOKENS (override the local window) · TWOB_NO_DIAGNOSTICS (skip post-edit checks) · TWOB_NO_LSP (regex symbol map instead of a language server) · TWOB_NO_SEATBELT / TWOB_SEATBELT=strict (relax / harden the run_command sandbox) · TWOB_NO_TRIM (keep bulky tool output in each request) · TWOB_NO_HISTORY (don't persist sessions) · TWOB_SUBAGENT_MODEL (run delegated sub-agents on a cheaper model) · TWOB_NO_UPDATE_CHECK (no background update check).

---

Honest caveats

• It reads and writes wherever you point it — with guardrails. 2B resolves absolute paths and paths outside the working directory on purpose; it's a personal tool for your machine. Writes are still confirmed in normal mode, plan mode refuses them entirely, reads of known-secret paths (~/.ssh, ~/.aws, …) prompt first, and on the cloud path run_command's writes are sandbox-confined to the workspace by default (see the sandbox bullet above). The command classifier is defense-in-depth, not an unbypassable boundary — the sandbox is the boundary.
• Switching to a stronger model mid-task hands it a tool-call history it didn't make. For these five simple tools that's low-risk (the wire format is unambiguously the new provider's own; only the choices inside came from a weaker model), but you may see mild "why did I read that file" moments. It is not the shim-degradation failure that sank the other harnesses.
• It's a full-screen TUI. That means it lives in the terminal's alternate screen, so a single mouse drag selects what's on screen, not scrolled-off history. For a classic inline REPL, run 2b --classic.

---

How it's built

• Python, standard library first. The only real dependencies are rich, prompt_toolkit, and textual. Every provider talks over urllib — no per-provider SDKs.
• A canonical conversation model that each provider serializes fresh on every request. That re-derivation is exactly what makes switching models mid-task safe.
• One worker thread per task, emitting events into a queue the UI thread drains and renders — so the tool code stays untouched off the main thread and one thread owns the terminal.
• The five-tool schema is frozen. It's what makes small models reliable, and it isn't up for redesign.

Built for local models, kept on task.