Documentation
¶
Overview ¶
Package causepolicy contains the provider-independent ALG eligibility policy for canonical terminal causes and bounded evidence.
Index ¶
Constants ¶
This section is empty.
Variables ¶
This section is empty.
Functions ¶
This section is empty.
Types ¶
type Eligibility ¶
type Eligibility string
Eligibility is the only authority-free action exposed by this policy.
const ( EligibilityStop Eligibility = "stop" EligibilityVerifier Eligibility = "verifier_eligible" )
type Reason ¶
type Reason string
Reason is a bounded explanation for the eligibility result.
const ( ReasonAuthoritative Reason = "authoritative_candidate" ReasonPreOutputTransport Reason = "pre_output_transport" ReasonMissingTrajectory Reason = "missing_trajectory" ReasonVerifierEligible Reason = "verifier_eligible" ReasonUnsafeAction Reason = "unsafe_action" ReasonInvalidInput Reason = "invalid_input" ReasonInsufficientEvidence Reason = "insufficient_evidence" ReasonOutputUncommitted Reason = "output_not_committed" ReasonUnsupportedCause Reason = "unsupported_cause" ReasonExplicitCompletion Reason = "explicit_completion" )
type Result ¶
type Result struct {
Eligibility Eligibility
Reason Reason
}
Result is a small value-only policy result. It grants no terminal, lifecycle, verifier, or recovery authority.
func Evaluate ¶
func Evaluate(input terminaldecision.Input) Result
Evaluate classifies one canonical terminal candidate. Invalid or incomplete evidence fails closed; only committed normal/transport/limit/provider-error candidates with a safe trajectory and complete tool facts may be offered to a verifier.
Source Files
Click to show internal directories.
Click to hide internal directories.