# unbrained: iterate fast.

steve preu, agentic systems architect (austria).

i design and build agent harnesses, memory and workflows, so the next session starts from what the last one learned.

## what i build

- **agent harnesses.** custom harnesses around the tools, permissions and habits your team already has.
- **memory and context.** durable, inspectable context that a fresh session can pick up, instead of one chat that grows forever.
- **deterministic workflows.** fixed steps where results must be repeatable, open room where an agent should investigate.
- **ready for agentic work.** starting from the goals, decisions and history you already have, in software teams and beyond.

from a first build to ongoing adaptation, preferably built together with your team.

## flagship: pm cli

project management is a context problem. pm cli keeps goals, acceptance criteria, decisions and history next to your code, as plain files that people and coding agents share. every change lands in an append-only history, so a fresh session can read what happened and carry on.

```sh
npm install -g @unbrained/pm-cli
```

recorded two-session handoff: https://unbrained.dev/pm-cli.md. source: https://github.com/unbraind/pm-cli. docs: https://pm-cli.unbrained.dev.

pm cli on the web:

- documentation: [pm-cli.unbrained.dev](https://pm-cli.unbrained.dev). quickstart, command reference, package catalog and changelog.
- browser: [pm-web.unbrained.dev](https://pm-web.unbrained.dev). boards, lists and item history in the browser, live for the whole team.
- chatgpt and codex: [pm-gpt.unbrained.dev](https://pm-gpt.unbrained.dev). read, plan and update hosted pm projects from chatgpt and codex.
- any mcp client: [pm-mcp.unbrained.dev](https://pm-mcp.unbrained.dev). a remote mcp endpoint with oauth for coding agents and ides.
- public search: [pm-search.unbrained.dev](https://pm-search.unbrained.dev). find decisions, tasks and fixes in the pm items of public repositories, pinned to the exact commit.

## shipped lately

- [pm-cli](https://github.com/unbraind/pm-cli): Agent-friendly, git-native project management for humans and coding agents. npm @unbrained/pm-cli@2026.10.18 (18 Oct 2026); source 96f3302 (18 Oct 2026).
- [pm-web](https://github.com/unbraind/pm-web#readme): Full web UI for pm-cli — browse, create, update, search and manage pm projects in the browser. Self-hosted via Docker or Node.js. pm catalog: official extension, install with `pm install npm:@unbrained/pm-web --project`. npm @unbrained/pm-web@2026.10.18 (18 Oct 2026); source 3ee03d1 (18 Oct 2026).
- [pm-graph](https://github.com/unbraind/pm-graph#readme): Knowledge graph and dependency graph extension for pm CLI workspaces, with optional Neo4j sync. pm catalog: official extension, install with `pm install npm:pm-graph --project`. npm pm-graph@2026.10.18 (18 Oct 2026); source a9175ac (18 Oct 2026).
- [pm-brief](https://github.com/unbraind/pm-brief#readme): Token-budgeted agent briefs and next-work plans for pm workspaces. pm catalog: official extension, install with `pm install npm:pm-brief --project`. npm pm-brief@2026.10.18 (18 Oct 2026); source b89eb22 (18 Oct 2026).
- [pm-gantt-chart](https://github.com/unbraind/pm-gantt-chart#readme): ASCII Gantt chart renderer for pm-cli. pm catalog: community extension, install with `pm install npm:pm-gantt-chart --project`. npm pm-gantt-chart@2026.10.18 (18 Oct 2026); source 3745247 (18 Oct 2026).

_pm catalog checked 18 Oct 2026, 20:57 UTC; npm checked 18 Oct 2026, 20:57 UTC; github checked 18 Oct 2026, 20:57 UTC._

all projects: https://unbrained.dev/projects.md

## how i work

- **start with what you have.** goals, decisions, notes and work traces are enough to begin, however messy. structure improves as we go.
- **write down what done means.** every task carries acceptance criteria, and every result is checked against them.
- **keep the reasoning.** decisions, rejected options and non-goals survive the rewrite, so nobody repeats the same detour.
- **you own the result.** your data and your workflow stay yours: free to adapt, extend or replace my tools, pm cli included.

## contact

hiring, or building something? either way, i would like to hear about it. tell me what you are working on and where agents could help.

open to engineering roles, consulting projects and collaborations.

- github: [github.com/unbraind](https://github.com/unbraind)
- x: [@stevepreu](https://x.com/StevePreu)

also on [linkedin](https://www.linkedin.com/in/stefanpreu/), [npm](https://www.npmjs.com/~unbrained), [tokscale](https://tokscale.ai/u/unbraind), [preu.at](https://preu.at).