A library of agent workflow definitions for deliberation and facilitation platforms.
Part of the Open Facilitation Library.
Workflows describe how AI agents facilitate group processes — how they get input from participants, process it, draw from memory/context, and produce outputs. Each workflow maps the concrete agent architecture of a platform or reusable pattern.
The key filter for inclusion: does an AI agent actively conduct, guide, or mediate a conversation or group process? This covers AI facilitation across domains: deliberation, market research, education, innovation workshops, dispute resolution, and more.
This complements OFL's facilitation patterns (abstract methods like cross-pollination or Delphi, described in the synthesis knowledge base) and agent skills (executable skills following the Agent Skills specification for use in platforms like Harmonica or OpenClaw).
schema/
workflow-schema.yaml # Schema definition
products/ # How real platforms implement AI facilitation
anthropic-interviewer.yaml
bot-mediation.yaml
complexchaos.yaml
crowdsmart.yaml
habermas-machine.yaml
harmonica.yaml
juno.yaml
listen.yaml
orchidea.yaml
outset.yaml
parlay.yaml
remesh.yaml
talk-to-the-city.yaml
thinkscape.yaml
Each workflow is a YAML file describing:
- Agents — what agent roles exist (facilitator, synthesizer, clustering engine, etc.)
- Participants — how people interact (text/voice/vote, sync/async, scale)
- Stages — the ordered pipeline: input → processing → memory → output
- Data flow — what goes in, what comes out, where data lives
- Pattern mapping — which OFL facilitation patterns it implements
- Source type — how the workflow was documented:
verified— from open-source code or detailed published methodologyresearched— from academic papers or detailed technical write-upsinferred— reconstructed from public marketing/website descriptions
See schema/workflow-schema.yaml for the full definition.
| Platform | Source | Type | Agents | Scale |
|---|---|---|---|---|
| Anthropic Interviewer | researched | Async 1-on-1 interviews | Interviewer + Analyst | 1,000+ |
| Bot Mediation | researched | AI dispute mediation | AI Bot Mediator | 2 (bilateral) |
| ComplexChaos | inferred | Async organizational alignment | Dialogue Agent + Pattern Synthesizer | 5-200 |
| CrowdSmart | researched | Generative Collective Intelligence | Collective Reasoning Agent + Pairwise Engine + Private LM | 10-10,000 |
| Habermas Machine | verified | Iterative consensus | Statement Generator + Critique Processor | 5-1,000 |
| Harmonica | verified | Async structured deliberation | Facilitator + Cross-Pollinator + Synthesizer | 3-500 |
| Juno | inferred | Unscripted AI interviews | Interviewer + Thematic Analyst | 5-1,000 |
| Listen | inferred | AI research pipeline | Recruiter + Interviewer + Analyst | 10-1,000 |
| Orchidea | researched | AI workshop facilitation | AI Ideator + Proposal Drafter | 3-500 |
| Outset | researched | Multimodal AI interviews | AI Moderator + Fraud Detector + Analyst | 10-5,000 |
| Parlay Ideas | researched | Classroom discussion facilitation | Parlay Genie + Participation Tracker | 5-40 |
| Remesh | inferred | Real-time collective dialogue | Clustering Engine + Consensus Predictor + Human Moderator | 20-5,000 |
| Talk to the City | verified | Elicitation + clustering | WhatsApp Bot + Topic Clusterer + Visualizer | 10-10,000 |
| Thinkscape | researched | Swarm intelligence | Conversational Surrogates + Orchestrator | 14-400 |
- Abstract workflows — reusable, product-agnostic workflow definitions extracted from the product layer (in
abstract/) - Schema.org JSON-LD — generate web pages with schema.org structured data from each YAML workflow, mapping to
HowTo/HowToStep(stages),SoftwareApplication(products),Role(agents), andDataset(outputs). This would make workflows discoverable by search engines and consumable by AI assistants (Claude, Perplexity, ChatGPT). Deliberation-specific concepts (cross-pollination, pairwise comparison, consensus) have no schema.org equivalent and would remain in the YAML schema. - OpenClaw integration — generate agent configurations for OpenClaw or similar orchestration frameworks from the abstract workflow layer
Add a new workflow by creating a YAML file following the schema. Product workflows go in products/, reusable abstract workflows in abstract/ (coming soon).
MIT