+
- 
+
- ▶ Watch the 26-min intro on YouTube — what it is, a walkthrough, and how to use it - + ▶ Watch the 26-min intro what it is, a walkthrough, and how to use it + |
+
+
+ 
+
+ + ▶ Watch the 2.5-hour build session a conversation → a shipped feature + |
+
@@ -240,7 +251,7 @@ uv run mypy roboco/
- [x] Database ORM (SQLAlchemy async)
- [x] Task lifecycle state machine
- [x] Multi-agent workspace management
-- [x] Agent prompts (20 agents)
+- [x] Agent prompts (22 agents)
- [x] Messaging API
- [x] Task API with full lifecycle
- [x] Git operations API
diff --git a/agents/prompts/base.md b/agents/prompts/base.md
index ccd5cbfa..3be21dac 100644
--- a/agents/prompts/base.md
+++ b/agents/prompts/base.md
@@ -1,6 +1,6 @@
# RoboCo Agent — Base
-You are an agent in **RoboCo**, an AI company with 20 AI agents + 1 human CEO. Your role-specific prompt names your verbs and your responsibilities; this file holds the rules every role obeys.
+You are an agent in **RoboCo**, an AI company with 22 AI agents + 1 human CEO. Your role-specific prompt names your verbs and your responsibilities; this file holds the rules every role obeys.
## Identity
diff --git a/docs/how-to.md b/docs/how-to.md
index 93f61586..12a9f939 100644
--- a/docs/how-to.md
+++ b/docs/how-to.md
@@ -2,7 +2,7 @@
-RoboCo is a virtual software company — 20 AI agents and one human: you. Not a
+RoboCo is a virtual software company — 22 AI agents and one human: you. Not a
swarm of bots, not a framework to wire together — an **organization**, with
roles, a chain of command, formal reviews, and sign-offs. You don't micromanage
it; you run it like a CEO. Drop work in at the top and the company carries it all
@@ -10,7 +10,7 @@ the way through planning, building, review, and documentation, then brings it
back to your desk for the final word. You act at the two ends; the organization
fills in everything between.
-What keeps twenty agents from dissolving into noise is that RoboCo is
+What keeps twenty-two agents from dissolving into noise is that RoboCo is
relentlessly opinionated about *how* work happens: everything is a task, no task
moves without acceptance criteria, and every task walks the same strict lifecycle
— built, QA'd, documented, PM-reviewed, approved — each step gated by role. The
@@ -260,7 +260,7 @@ for a demo — it's a real page RoboCo's agents shipped to RoboCo's own control
panel. A company building its own product, in front of you, is the whole point of
RoboCo. What makes that hold together isn't a clever model or a lucky run; it's
the **organization** — the roles, the gated lifecycle, the reviews and the
-sign-offs that keep twenty agents moving as a company instead of a crowd. Run
+sign-offs that keep twenty-two agents moving as a company instead of a crowd. Run
as many of these passes as you like, across as many projects as you like.
---
diff --git a/docs/rag/architecture/org-structure.md b/docs/rag/architecture/org-structure.md
index 6a9dc18b..a8834949 100644
--- a/docs/rag/architecture/org-structure.md
+++ b/docs/rag/architecture/org-structure.md
@@ -5,38 +5,43 @@
```
CEO (Renzo - Human)
|
- +-- Board (3 agents)
- +-- Product Owner
- +-- Head of Marketing
- +-- Auditor (silent observer)
- |
- +-- Main PM
- |
- +-- Backend Cell
- +-- Frontend Cell
- +-- UX/UI Cell
+ +-- Board (4 agents)
+ | +-- Product Owner
+ | +-- Head of Marketing
+ | +-- Auditor (silent observer)
+ | +-- PR Reviewer (read-only safety check on PRs)
+ |
+ +-- Main PM
+ |
+ +-- Backend Cell
+ +-- Frontend Cell
+ +-- UX/UI Cell
```
## Agent Count
| Role | Count |
|------|-------|
-| CEO | 1 (human) |
+| CEO | 1 (human — not counted in the 22) |
| Product Owner | 1 |
| Head of Marketing | 1 |
| Auditor | 1 |
+| PR Reviewer | 1 |
| Main PM | 1 |
| Cell PMs | 3 |
| Developers | 6 (2 per cell) |
| QAs | 3 (1 per cell) |
| Documenters | 3 (1 per cell) |
-| **Total** | **20** (19 AI + 1 human) |
+| Prompter (Intake) | 1 (on-demand) |
+| Secretary | 1 (on-demand) |
+| **Total AI agents** | **22** |
## On-Demand Roles (Human-Facing)
-Two roles sit outside the standing delivery org above. They are **human-only**,
-**spawned on demand** as live chat sessions, and are not part of the 20-agent
-count:
+Two of the 22 — the Prompter (Intake) and the Secretary — sit outside the
+standing delivery org above. They are **human-only** and **spawned on demand**
+as live chat sessions; they are counted among the 22, but unlike the standing
+org they have no lifecycle verbs and no outward agent comms:
| Role | Purpose |
|------|---------|
@@ -59,7 +64,7 @@ Neither has lifecycle verbs or outward agent comms. See
| Team | Members |
|------|---------|
| executive | ceo |
-| board | product-owner, head-marketing, auditor |
+| board | product-owner, head-marketing, auditor, pr-reviewer-1 |
| management | main-pm, be-pm, fe-pm, ux-pm |
| developers | all devs |
| qa | all QAs |
@@ -80,6 +85,7 @@ Each role can communicate with:
| CEO | Everyone |
| Board | CEO, other board, Main PM |
| Auditor | Everyone (silent read all) |
+| PR Reviewer | Read-only; posts one change-request on the PR itself, no agent comms |
| Main PM | CEO, Board, Cell PMs |
| Cell PM | Main PM, cell members |
| Cell Members | Cell PM, other cell members |
diff --git a/usage.md b/usage.md
index ac9e41b2..64ebf951 100644
--- a/usage.md
+++ b/usage.md
@@ -4,7 +4,7 @@ Operating the AI company after deployment.
## The Organization
-20 AI agents organized as a company:
+22 AI agents organized as a company:
```
CEO (You)
@@ -235,12 +235,19 @@ docker logs -f roboco-agent-be-dev-1
### Resource Usage
-Each agent container uses ~500MB-2GB RAM depending on context. With 128GB RAM:
-- 3 agents: ~6GB
-- 6 agents: ~12GB
-- 20 agents: ~40GB (the intake interviewer is on-demand — it only runs while you're drafting a task)
+RAM is modest. Agent containers are spawned on demand and torn down when
+their work is done, so you rarely have more than a handful live at once — and
+the on-demand Intake and Secretary only run while you're interacting with them.
+Steady-state memory is dominated by the standing services (Postgres, Redis,
+and especially Ollama with its models loaded), not by the agents.
+
+Storage is the larger footprint: the built (or pulled) image set. The agent
+images all share a common base layer, so on disk they cost far less than their
+nominal sizes added together. `docker system prune` reclaims old image versions
+and stopped agent containers.
Monitor with:
```bash
-docker stats --filter "name=roboco-agent"
+docker stats # live RAM / CPU per running container
+docker system df # image / container / build-cache disk usage
```