Workflows
Workflows
This chapter defines the core workflows used across the FactHarbor system.
Each workflow describes:
- Purpose
- Participants
- Steps
- Automation vs. manual work
- (Wherever applicable) linear ASCII flow: a → b → c → d
Workflows included:
- Claim Workflow
2. Scenario Workflow
3. Evidence Workflow
4. Verdict Workflow
5. Re-evaluation Workflow
6. Federation Synchronization Workflow
7. User Role & Review Workflow
8. AKEL Workflow
9. Global Trigger Flow
10. Entity Lifecycle Notes
Claim Workflow
Purpose:
Transform raw text or input material into a normalized, classified, deduplicated, and versioned claim ready for scenario evaluation.
Participants:
- Contributor
- AKEL
- Reviewer
Steps
- Ingestion
- User submits text, URL, transcript, or multi-claim content
- AKEL extracts one or multiple claims
2. Normalization
- Standardizes wording
- Reduces ambiguity
- Flags implicit assumptions
3. Classification (AKEL draft → Reviewer confirm)
- ClaimType
- Domain
- Evaluability
- SafetyCategory
4. Duplicate & Similarity Detection
- Embeddings created
- Similar claims found
- Reviewer merges, splits, or confirms uniqueness
5. Version Creation
- New ClaimVersion stored
- Every edit creates a new immutable version
6. Cluster Assignment
- AKEL proposes cluster membership
- Reviewer confirms
7. Scenario Linking (optional)
- Existing scenarios connected
- AKEL may propose new drafts
8. Publication
- Claim becomes active and visible
Flow:
Ingest → Normalize → Classify → Deduplicate → Cluster → Version → Publish
Scenario Workflow
Purpose:
Define the specific analytic contexts needed to evaluate each claim.
Participants:
- Contributor
- Reviewer
- Domain Expert
- AKEL
Steps
- Scenario Proposal
- Drafted by contributor or generated by AKEL
2. Completion of Required Fields
Must include:
- Definitions
- Assumptions
- ContextBoundary
- EvaluationMethod
- SafetyClass
- VersionMetadata
3. Safety Interception (AKEL)
Flags:
- non-falsifiable structures
- pseudoscientific assumptions
- unsafe contexts
4. Redundancy & Conflict Check
- Similar scenarios merged
- Contradictions flagged
5. Reviewer Validation
Ensures clarity, neutrality, and methodological validity.
6. Expert Approval (mandatory for high-risk domains)
7. Version Storage
- Each revision = new ScenarioVersion
Flow:
Draft → Validate → Safety Check → Review → Expert Approval → Version → Activate
Evidence Workflow
Purpose:
Structure, classify, validate, version, and link evidence to scenarios.
Participants:
- Contributor
- Reviewer
- Domain Expert (when needed)
- AKEL
Steps
- Evidence Submission
- File, dataset, URL, or extracted text
2. Metadata Extraction (AKEL)
- EvidenceType
- Category
- Provenance
- Study design
- ExtractionMethod
- ReliabilityHints
3. Relevance Check
Reviewer verifies which scenarios the evidence applies to.
4. Reliability Assessment
- AKEL proposes reliability
- Reviewer confirms
- Expert review for complex papers
5. ScenarioEvidenceLink Creation
Each link stores:
- relevance score
- justification
- evidence version
6. Versioning
- Any update = new EvidenceVersion
Flow:
Submit → Extract Metadata → Evaluate Relevance → Score Reliability → Link → Version
Verdict Workflow
Purpose:
Generate likelihood estimates per scenario based on evidence and scenario structure.
Participants:
- AKEL (drafts)
- Reviewer
- Domain Expert
Steps
- Evidence Aggregation
Collect relevant evidence versions.
- Evidence Aggregation
2. Draft Verdict Generation (AKEL)
Outputs:
- likelihood range
- uncertainty factors
- conflict detection
- sensitivity analysis
3. Reasoning Draft
Structured explanation chain generated by AKEL.
4. Reviewer Validation
Ensures logic, evidence fit, no hallucinations.
5. Expert Review
Required for:
- medicine
- psychology
- engineering
- political misinformation
- controversial or risky domains
6. Verdict Storage
- Every update creates a new VerdictVersion
Flow:
Aggregate → Draft Verdict → Draft Explanation → Review → Expert Approval → Version
Re-evaluation Workflow
Purpose:
Keep verdicts current when evidence or scenarios change.
Trigger Types
- Evidence updated, disputed, retracted
- Scenario assumptions changed
- Claim reclassification
- AKEL contradiction detection
- Federation sync
Steps
- Trigger Detection
Re-evaluation engine receives event.
- Trigger Detection
2. Impact Analysis
Find affected:
- scenarios
- evidence links
- verdicts
3. AKEL Draft Re-calculation
New:
- likelihood
- reasoning
- uncertainty
4. Reviewer Validation
5. Expert Review (high-risk)
6. Version Storage
Flow:
Trigger → Analyze → Recompute → Review → Expert → Version
Federation Synchronization Workflow
Purpose:
Exchange structured data between nodes.
Steps
- Detect version changes
- Build bundle (diff + Merkle tree + signatures)
- Push to peers
- Validate lineage + hashes
- Resolve conflicts (merge or branch)
- Optional re-evaluation
Flow:
Change → Bundle → Push → Validate → Merge/Fork → Update
User Role & Review Workflow
Purpose:
Ensure correctness, neutrality, safety, and resistance to manipulation.
Steps
- Submission
Claim / scenario / evidence / verdict.
- Submission
2. Auto-check (AKEL)
Flags unsafe content, contradictions, format issues.
3. Reviewer Validation
4. Expert Validation
Required for sensitive domains.
5. Moderator Oversight
Triggered by suspicious behavior.
Flow:
Submit → Auto-check → Review → Expert → Moderator (if needed)
AKEL Workflow
Purpose:
Support extraction, drafting, structuring, and contradiction detection.
Stages
A — Input Understanding:
Extraction, normalization, classification.
B — Scenario Drafting:
Definitions, boundaries, assumptions.
C — Evidence Processing:
Retrieval, summarization, ranking.
D — Verdict Drafting:
Likelihood, explanations, uncertainties.
E — Safety & Integrity:
Contradictions, hallucination detection.
F — Human Approval:
Reviewer and/or expert.
Flow:
Input → Drafts → Integrity → Human Approval
Global Trigger Flow (Cascade)
Trigger Sources:
- Claim change
- Scenario change
- Evidence change
- Verdict contradiction
- Federation update
- AKEL model improvements
Cascade Flow:
Trigger → Dependency Graph → Re-evaluation → Updated Verdicts