Claude works 24/7. User sets goals, not tasks. Quality over speed.
Evolved from 6 actors to 3 tiers. Orchestrator subsumes Creator, Dispatcher, and Optimizer.
Sets goals directly with Orchestrator
1 persistent instance
Up to 10 ephemeral
5-6 persistent — launched on boot, recycled with /clear
Investigation
HTML, CSS, UI
Deploy, SSH
Schema, SQL
Email domain
How the Orchestrator builds ~40-line prompts that turn generic Raw Claudes into specialists.
Source: bespokes table in Master_Claude.db
Task-specific framing, not generic roles. Tells the Claude what kind of worker it is for this task.
The actual work — prominent and clear
specialist_contexts tableOnly 2-3 relevant skills, not everything
focused_claude_mcps table27 bespokes with multiple scopes in the DB. The prompt builder selects the right identity + context + tools per task. A Raw Claude receiving this prompt becomes the specialist it needs to be.
Visual flow from task to ready-to-send prompt.
From user or queued work
Invokes prompt_builder.py
bespokes → identityspecialist_contexts → domain knowledge (with scope)focused_claude_mcps → available toolslearnings_fts → relevant existing knowledge~40 lines, focused and minimal
Ephemeral task worker
Persistent domain expert
Three layers of context injection, in priority order. Key insight: on-demand context > hints at action > pre-loaded docs.
| Layer | Mechanism | When It Fires | Token Cost |
|---|---|---|---|
| 1 | MCP on-demand docs (Context7) | Claude pulls what it needs | Zero until used |
| 2 | Contextual hints (tool_hints.json via hooks) | At moment of action | ~5 lines |
| 3 | Lightweight skills (SKILL.md with context: fork) | When keyword triggers | Fork context |
This replaces pre-loading massive CLAUDE.md files. Each layer fires only when needed, keeping base context lean. The prompt builder uses layer 1 (MCP lookups) + layer 3 (skill triggers) when assembling prompts for Raw Claudes.
6 persistent Claudes launched on startup. Raw Claudes spawned on-demand.
| Order | Claude | Why Persistent |
|---|---|---|
| 1 | Orchestrator | Core — receives goals, builds prompts, validates |
| 2 | Researcher | Always-on investigation |
| 3 | Web Designer | Frequent UI work |
| 4 | Cloudflare/Infra | SSH, deployment |
| 5 | Database | Schema coordination |
| 6 | Separate domain |
Messaging API enables any-to-any communication. Replaced file-based inbox system.
POST /send-prompt/{id}
POST /complete/{id}
GET /messages/recent
| Feature | Description | Status |
|---|---|---|
| Send Prompt | POST to /send-prompt/{session_id} | Working |
| Complete Signal | POST to /complete/{session_id} | Working |
| Sessions API | /sessions/active lists all running Claudes | Working |
| Message History | /messages/recent for audit trail | Working |
| Command Center | Web UI for /chat, /messages, /sessions | Working |
| State | Where | Who Acts |
|---|---|---|
| queued | Orchestrator queue | Orchestrator dispatches when Raw Claude available |
| prompt_built | /build_prompt output | 3-part prompt assembled from DB |
| in_progress | Raw Claude / Specialist | Worker actively executing |
| verifying | Orchestrator | Orchestrator validates results |
| complete | Logged | /clear sent, Raw Claude recycled |
Different validation levels based on task complexity. Orchestrator handles all review paths.
| Complexity | Validation | Human Review |
|---|---|---|
| Simple | Quick keyword match | None — auto-proceed |
| Complex | Git evidence, file changes | Orchestrator spot-check |
| Critical | Deep verification criteria | User approval required |
| Component | Status | Notes |
|---|---|---|
| Launcher API (localhost:5111) | Working | /launch, /send-prompt, /complete with validation |
| Sessions tracking | Working | Token fields, worker_type, learnings |
| Workers registry (workers.db) | Working | Named workers → active sessions, short_name for tabs |
| Idle detection | Working | worker_idle_detector.py reads status line |
| Validator | Working | validator/validator.py — catches fake completions |
| autonomous_supervisor.py | Working | Launches workers, monitors health |
| Worker completion protocol | Working | Orchestrator feedback loop validates before recycle |
| /build_prompt skill | Working | Orchestrator prompt building from DB context |
| prompt_builder.py | Working | DB context lookup for 3-part prompt assembly |
| specialist_contexts table | Working | Scoped domain knowledge (27 bespokes, multiple scopes) |
| Superseded | Replaced by messaging API | |
| Superseded | Orchestrator handles dispatch directly | |
| Superseded | Quality review folded into Orchestrator |
Orchestrator monitors for conditions that need intervention.
| Trigger | Detection | Action |
|---|---|---|
| Idle pool + no queued work | Raw Claudes available, nothing to send | Create proactive tasks or notify user |
| High error rates | > 3 failures in 1 hour for same task type | Pause, investigate, adjust prompt |
| Stale tasks | Queued > 24 hours without progress | Review blockers, reassign or escalate |
| Token thresholds | Session approaching 80% context limit | Preemptive /clear with handoff.md |
| Loop detection | Same task sent > 3 times | Analyze patterns, fix root cause |
| Quality drift | Rising verification failure rate | Adjust prompt templates, update bespokes |
Replace 30 terminal windows with one web conversation.
Watching 30 terminals, losing context
ORCHESTRATOR + POOL
(3-tier architecture)
Talks only to Orchestrator