Examples
Examples
Examples teach how governed operational decisions move through the runtime, not how to wire a single adapter. Each pattern names who may transition state, what evidence is required, and where humans intervene — before any side effect reaches an operational system.
→ Runtime Taxonomy · Architecture · Loop Engine vs Workflow Engines
Canonical runtime flow
Every priority pattern on this page follows the same spine:
1Provider (intelligence)2 ↓3Loop + Guards (governance)4 ↓5Channel (human surface)6 ↓7Integration (operational system)8 ↓9Evidence (audit + learning)Providers (intelligence systems) recommend or classify — they do not commit transitions alone. Guards enforce deterministic policy before state changes. Channels (human coordination) are where operators approve, reject, or escalate — not where CRM writes occur. Integrations (systems of record) persist state and execute side effects after approval. Evidence explains why each transition was allowed.
| Layer | Role in examples | | --- | --- | | Provider | Score, draft, classify | | Channel | Slack, Teams, doc review | | Integration | Salesforce, Sheets apply, PagerDuty, MAP |
What examples demonstrate
| Principle | In practice |
| --- | --- |
| Governance | human-only, confidence-threshold, evidence-required guards block unauthorized transitions |
| Evidence | Structured fields on every material transition — not reconstructed logs |
| Deterministic boundaries | Guards are runtime checks, not prompt instructions |
| Human escalation | Low confidence or high risk → PENDING_* states routed to a Channel |
| AI assistance | Providers submit recommendations inside the loop — never self-approve |
ABM and RevOps operational AI
GTM and revenue operations are where operational AI needs runtime governance: models draft and score, humans decide in Slack, systems of record update after policy passes.
SDR qualification loop
Provider scores inbound lead → Slack escalation → SDR approves → CRM updated.
Layers: Provider · Channel (Slack) · Integration (CRM)
Proposal approval loop
AI drafts in Google Docs → sales/legal approve in Slack → Salesforce advances.
Dual-surface: Docs (work) + Slack (decisions)
Campaign approval loop
Outbound copy generation → brand/compliance guards → Slack → MAP executes.
Pricing exception loop
Sheets staging → finance approval in Slack → ERP/CRM apply with evidence.
Dual-surface and platform patterns
Google Docs + Slack
AI-generated proposal in Docs → Slack approval → governed transition → apply path.
Flow: Provider → Loop + Guards → Channel (Slack) → Integration (Docs/Sheets)
Surface: Loop Engine Cloud · Narrative pattern
Google Sheets + Slack
Spreadsheet staging → human approval in Slack → gated apply to Sheets.
Flow: Provider (optional) → Loop + Guards → Channel → Integration
Surface: Loop Engine Cloud · Narrative pattern
PagerDuty incident loop
AI classification + guarded escalation + PagerDuty alerting with audit trail.
Flow: Provider → Loop + Guards → Channel → Integration (PagerDuty)
Surface: OSS adapters · integration guide
Workflow + Loop
Temporal or Salesforce durable steps with Loop Engine governing transitions.
Flow: Provider → Loop + Guards → Channel → Integration (workflow activity)
Surface: Narrative · complements workflow engines
OSS runnable patterns
1git clone https:"cmt">//github.com/loopengine/loop-examples2cd loop-examples && pnpm installAI Replenishment (Claude)
Provider recommendation → confidence guard → human approval → order integration.
Runnable: Yes · loop-examples/ai-actors/claude
Infrastructure Change Approval
Provider blast-radius analysis → SRE Channel → rollout Integration.
Runnable: Coming soon
Fraud Review
Provider scoring → analyst Channel → case disposition with evidence.
Runnable: Coming soon
Expense Approval
Loop + Guards + human Channel — governance baseline without an AI Provider.
Runnable: Yes · loop-examples/expense-approval
OpenClaw Channel
Messenger Channel routing for PENDING_HUMAN_APPROVAL — OSS adapter.
Runnable: Pattern doc · see OpenClaw integration
Commerce Gateway
Governed tool execution — Integration after policy passes (not a Provider).
Runnable: See integration + examples repo index
Supporting patterns
| Pattern | Focus |
| --- | --- |
| Demand Signal | Signals → loop start (no Provider) |
| Postgres Persistence | Integration (LoopStore) swap |
| Event Streaming | Evidence via event timeline |
| AI Replenishment (OpenAI) | Same flow as Claude — adapter swap |