Response Schema

The canonical output format.

The BaseCred Decision Engine returns a DecisionOutput object. This schema is strict and will not change without a major version bump.

JSON Structure

{
  "decision": "ALLOW",
  "confidence": "HIGH",
  "explain": [
    "String explanation of why the decision was made",
    "Multiple reasons may be provided"
  ],
  "constraints": [],
  "retryAfter": null,
  "ruleIds": ["rule_id_1", "rule_id_2"],
  "version": "1.0.0"
}

Field Definitions

Field	Type	Description
decision	Enum	The final recommendation. One of ALLOW, DENY, or ALLOW_WITH_LIMITS.
confidence	Enum	Qualitative certainty: LOW, MEDIUM, HIGH, or VERY_HIGH. Derived from signal coverage and consistency.
explain	string[]	Human-readable reasons. Safe to display to end-users.
constraints	string[]	Limitations when decision is ALLOW_WITH_LIMITS. Empty array for ALLOW or DENY.
retryAfter	number | null	Seconds to wait before retrying, for rate-limited or temporary denials. null if not applicable.
ruleIds	string[]	IDs of rules that contributed to this decision. Useful for debugging and auditing.
version	string	Engine version used to produce this decision (SemVer).

Optional Fields

These fields are populated by the progression layer and are not part of the core decision output:

Field	Type	Description
accessStatus	Enum | undefined	Retail-facing access status: eligible, limited, not_ready, or blocked.
blockingFactors	string[] | undefined	High-level, fixable guidance list. Does not expose raw scores.

Decision Values

Value	Meaning
ALLOW	Subject meets all requirements for this context.
DENY	Subject does not meet requirements. Check explain for reasons.
ALLOW_WITH_LIMITS	Subject is allowed with restrictions. Check constraints for details.

Confidence Tiers

Tier	Meaning
LOW	Few signals available. Decision is a best-effort estimate.
MEDIUM	Partial signal coverage. Decision is reasonably reliable.
HIGH	Good signal coverage. Decision is reliable.
VERY_HIGH	Full or near-full signal coverage. Decision is highly reliable.