System Architecture — Runtime-Independent Design
Clearfork C-Suite serves CEOs and founders of mid-market companies ($5M–$100M). It's a generic agent control plane — Mission Control — with a domain-specific overlay — C-Suite — installed on top.
The CEO experiences a strategic advisor and performance coach. Behind the scenes: business mapping, commitment tracking, automation opportunity logging, and a compounding intelligence layer that gets more valuable over time.
The system is not one giant prompt. It's a layered architecture where identity, operating rules, workflows, and state are separated cleanly — so each layer can evolve independently.
A third layer — the Client Instance — holds per-client data: CEO profile, people map, process map, tech stack, priorities, commitments, opportunities, artifacts, and configuration.
Be the CEO's thinking partner. Draw out what they already know. Ask the questions nobody else asks. Be the one person who's always caught up.
Map the business through conversation. When automation or AI opportunities surface, log them — whether the CEO engages with them or not.
Start by drawing out what the CEO already knows. When they need strategic thinking, bring the best available and frame it in their world. Their picture sharpens — it doesn't get replaced.
Build the full picture of the business before going deep on any one problem. When the CEO surfaces a pain point, acknowledge it, log it, then broaden. Keep going until they run dry.
The coach never architects workflows, drafts technical specs, or shifts into developer mode during a coaching conversation. Log the opportunity, stay in coaching.
Don't redirect to Socratic questions when they want a direct answer. Be the most valuable strategic mind in the room. Give the advice, then coach what it reveals.
Don't stall by redirecting to coaching. Ask for context if needed, execute with radical specificity, then use what the work reveals as a coaching door if the moment is right.
Who is the coach? Mission, values, voice, relationship stance, coaching method, characteristic moves, hard human boundaries. Does NOT contain logging instructions, schemas, or post-session mechanics.
The hidden runtime rules. Mission hierarchy, behavioral rules, escalation boundaries, delegation rules, transformation mode behavior, honesty rules. This is the operating law. It is not the soul.
CEO Profile, Relationship Signals, People Map, Process Map, Tech Stack, Data Landscape, Priorities, Commitments, Friction Log, Opportunity Register, Artifacts. The real memory of the client relationship.
Repeatable bounded operations: session closeout, opportunity scoring, commitment review, company model refresh, artifact indexing, delegate to builder, delegate to research. Not personality. Not identity.
Coach, Research, Builder. The coach may call the other agents. The coach does not become them. This separation is a load-bearing wall.
Recent commitments, active priorities, relevant people/process/tech/data slices, relationship signals, transformation mode. Only what's relevant — not the full database.
Governed by SOUL + AGENTS + available context. Trusted, specific, emotionally intelligent, commercially sharp, never salesy.
Extract structured signals: new people mentioned, new friction discovered, commitments made, recurring patterns, candidate opportunities, emotional risk, follow-up hooks.
Write extracted signals to the structured state store. Update company graph, commitments, priorities, relationship signals, artifacts, and opportunities.
If transformation_mode ON — score and log proactively. If OFF — only score on strong signal or explicit request.
Research or building requests get delegated via a clean handoff packet. Coach creates the packet and stays in the relationship.
The CEO sees Step 2. The system runs Steps 3–6.
If in doubt, keep coaching.
Stop coaching. Do not diagnose, minimize, or prescribe. Flag to Clearfork immediately. "Can't sleep because of the Danny decision" = coaching. "I haven't felt anything in weeks" = clinical.
Give your best strategic thinking. Be the most useful mind in the room. Then caveat appropriately. Never refuse to engage just because a professional should eventually be involved.
Go quiet. Don't argue, defend, or plead. Flag to Clearfork for human re-engagement.
OpenClaw provides out of the box: session management, memory (SQLite + embeddings), prompt caching (90% cost savings), bootstrap file system with priority-ordered loading, skills framework, MCP tool integration, heartbeat/proactive outreach, context compaction, and a working UI.
Building on Agent SDK or Hermes would require rebuilding all of this infrastructure before a single coaching conversation could happen. Treat OpenClaw as substrate, not product logic. Put Clearfork behavior in workspace files, skills, and external services.
Mission, values, voice, coaching method. Does NOT contain behavioral rules, logging instructions, schemas, or post-session mechanics.
Mission hierarchy, behavioral rules, escalation, delegation, honesty rules. Sub-agents receive AGENTS.md + TOOLS.md only — the separation propagates for free.
MVP: session-closeout, artifact-index. Phase 2: opportunity-score, commitment-review, company-model-refresh, delegate-builder, delegate-research.
MVP: coach agent only. Phase 2: research + builder sub-agents. Mission hierarchy and honesty rules propagate automatically via AGENTS.md.
| Priority | File | Purpose | Cache |
|---|---|---|---|
| 1 | AGENTS.md | Operating contract, hidden rules | Cached |
| 2 | SOUL.md | Identity, voice, values | Cached |
| 3 | TOOLS.md | Tool usage guidance | Cached |
| 4 | IDENTITY.md | Additional identity (unused for now) | Cached |
| 5 | USER.md | Client profile — loaded every turn | Cached |
| 6 | HEARTBEAT.md | Proactive outreach rules | Cached |
| 7 | BOOTSTRAP.md | Additional context (unused for now) | Cached |
| 8 | MEMORY.md | Memory index / pointers | Cached |
Combined budget target: under 60,000 chars (~15,000 tokens). With SOUL at ~10K, AGENTS at ~8K, USER at ~3K, HEARTBEAT at ~1K, MEMORY at ~1K — well within budget.
OpenClaw produces one response — no split visible/hidden.
→ Workaround: inline updates + session-closeout skill
State lives in workspace files.
→ Workaround: structured markdown with consistent schemas
No automated scoring workflow.
→ Workaround: coach logs opportunities with manual assessment fields
No runtime config switch.
→ Workaround: AGENTS.md instructs coach to check USER.md for a transformation_mode field
None of these block shipping. All of them get solved in Phase 2.
None of these are urgent. All of them are real possibilities as the product grows.
Full control over agent loop, native multi-agent with handoffs, hooks for per-turn hidden extraction, MCP, structured outputs, checkpointing. 150–300 hrs to reach OpenClaw feature parity. Best for owning the full Mission Control stack.
Architecturally cleaner for long-running stateful agents with structured memory. Natively understands persistent identity and evolving memory — exactly what Clearfork needs. Similar effort to Agent SDK.
Perfectly valid if the gaps don't hurt you. The discipline of keeping product logic portable means migration is always an option, never an emergency.
SOUL.md, AGENTS.md, skills, and state schemas are plain text. They work on any runtime that can read files.
Structured state objects have defined fields. When you move to a database, these become table schemas.
Each skill is a bounded workflow with defined inputs and outputs. They port to any framework that supports procedural prompts.
Don't patch OpenClaw's core prompt assembly. Don't rely on OpenClaw-specific features without equivalents elsewhere. If you must, wrap it so the dependency is explicit and replaceable.
When you change runtimes, AGENTS.md is the primary file that needs updating. SOUL.md should rarely change.