Built-in examples
OpenExpertise ships with 12 examples covering every node kind, every fan-out pattern, and three real cognitive workflows (deep research, systematic debugging, brainstorming).
Each example has a README.md, an experience.yaml, fixtures, and a mocked-LLM e2e test — no API key needed to validate the structure.
Hero examples
Review Branch ★ : The hero demo. Multi-dim code review + verifier + score + advisor evolution. Three reviewers (bugs/perf/tests) fan out, find issues, miss SQL injection — advisor proposes adding a security dimension. : Nodes: tool + agent ×3
Tri-CLI Orchestration ★ : Claude → Codex → Gemini in one DAG. The headline cross-vendor demo. : Nodes: cli-agent ×3
Deep Research : Multi-vendor research: Claude Code WebSearch + Gemini Google Search → cited synthesis. : Nodes: tool + agent + cli-agent ×2
Systematic Debugging : The superpowers systematic-debugging skill as a YAML flow. Hypothesize → verify → fix via Claude Code → re-test. : Nodes: tool ×2 + agent + cli-agent
Brainstorming : The superpowers brainstorming skill: diverge across 3 angles → cluster → critique → synthesize top 3. : Nodes: tool + agent ×2 + cli-agent ×2
Primitive demos (one node kind at a time)
Hello Tool — Smallest possible flow. No LLM. tool only.
Dataset Aggregate — Load CSV → aggregate. dataset + tool.
Agent Echo — Single agent with structured output. agent.
Oncall Runbook — Investigate an incident across 3 dimensions via for_each fan-out.
Issue Triage — Classify → search dupes → conditional dedup → route. Shows when: edges.
Release Gates — License + changelog + coverage + Claude-Code security scan → release gate.
CLI Orchestration — Two providers: Claude Code summarizes, Codex critiques.
All examples at a glance
| Example | Featuring | LLM required? | CLIs required? |
|---|---|---|---|
| hello-tool | tool node | no | no |
| dataset-aggregate | dataset + tool | no | no |
| agent-echo | agent node | yes | no |
| review-branch ★ | for_each, when:, array_append, multi-phase | yes | no |
| oncall-runbook | for_each fan-out, synthesis pipeline | yes | no |
| issue-triage | when: conditional, skip-cascade | yes | no |
| release-gates | tool + cli-agent + agent mix, fan-in | yes | claude |
| cli-orchestration | cli-agent chain, 2 providers | no | claude, codex |
| tri-cli-orchestration ★ | 3-provider CLI chain | no | claude, codex, gemini |
| deep-research | dual for_each, fan-in, multi-vendor search | yes | claude, gemini |
| systematic-debugging | cli-agent for_each, 6-phase fix loop | yes | claude |
| brainstorming | cli-agent fan-out ×2, cluster + synthesize | yes | claude |
Running any example
# Validate YAML schema (no LLM or CLIs needed)
oe validate examples/<name>
# Run with live dashboard
oe run examples/<name> --tui
# Run with args
oe run examples/<name> --args '{"key":"value"}'
# Inspect state after run
oe state <field>
# Full event log
oe inspect <run-id>
# Trigger evolution advisor
oe evolve <run-id>Want to share your own experience? See Sharing experiences (coming soon).