blazalek/open-smtp-error-dataset

Open SMTP Error Dataset

Open SMTP Error Dataset is an English-language, machine-readable reference package for SMTP enhanced-status knowledge and cautious operational classification. Version 1.1.0 contains 92 stable knowledge records and a separate, auditable catalog of 120 classification rules.

The package is a reference artifact, not a live provider-policy feed. It helps with observability, support, parser testing, and bounded delivery operations; it does not establish recipient state or final inbox placement.

Configurations and contracts

The default configuration is the knowledge dataset. The classification_rules configuration is a distinct rule catalog for applications that need an explicit decision order. Do not treat either configuration as a claim about a current provider policy.

Knowledge classification fields

Each knowledge record includes these operational_classification fields:

• —rule_key
• —failure_persistence
• —cause_family
• —recipient_validity
• —retry_disposition
• —suppression_disposition
• —confidence
• —needs_review

The first matching classification rule is represented by rulekey. A needsreview value of true means the result is suitable for human review or additional context gathering, not an irreversible automated action.

Classification-rule schema

Each row in the classification_rules configuration has these public fields:

• —schema_version
• —rule_key
• —source
• —precedence_tier
• —enhancedstatuscode
• —enhancedstatusprefix
• —match_field
• —match_operator
• —match_value
• —failure_persistence
• —cause_family
• —recipient_validity
• —retry_disposition
• —suppression_disposition
• —confidence
• —needs_review
• —rationale
• —source_title
• —evidence_reference
• —knowledgerecordid
• —excludedenhancedstatus_codes

Rule evaluation is deterministic: exact mappings are considered first, exclusions are honored before fallbacks, then cautious fallback rules are considered. Results outside that coverage remain unclassified. In particular, 5.7.28 is intentionally unclassified by policy, while the exact mapping for 4.7.28 remains available.

Load both configurations

~~~python from datasets import load_dataset

knowledge = loaddataset( "blazalek/open-smtp-error-dataset", split="train", ) rules = loaddataset( "blazalek/open-smtp-error-dataset", "classification_rules", split="train", ) ~~~

Use the knowledge configuration for explanatory catalog content. Use the rule configuration when an application needs to show its rule source, precedence, confidence, and review requirement.

Public boundary and responsible use

Raw SMTP responses, observed delivery-event examples, recipient information, personal data, provider-specific wire text, and internal working material are excluded from this release. The public package contains privacy-safe structured knowledge, rules, and provenance only.

Do not use a text-only signal or a result marked needs_review to automatically suppress a recipient, close an investigation, or make another irreversible delivery decision. Keep operational telemetry in the system authorized to hold it.

Public files

The proposed local v1.1.0 base payload contains:

• —README.md
• —bounce-recognition-guide.md
• —CHANGELOG.md
• —NOTICE.md
• —data/open-smtp-error-knowledge.jsonl
• —data/open-smtp-error-knowledge.csv
• —data/open-smtp-error-knowledge.parquet
• —data/bounce-recognition-rules.json
• —data/dataset-manifest.json
• —data/dataset.schema.json
• —data/classification-rules.jsonl
• —data/classification-rules.csv
• —data/classification-rules.parquet
• —data/classification-rule.schema.json

LICENSE.md is deliberately absent until its identifier and treatment receive a separate owner decision. The NOTICE file records third-party attribution; it does not state a licence for the overall dataset.

See CHANGELOG.md for version history and bounce-recognition-guide.md for the recognizer's separate integration contract.