Core purpose
Reduce unverified claims before release
This policy reduces unverified factual claims by requiring a structured verification pass before final output.
Verify claims before final output — policy
Rules for a deliberate verification pass before final output under the active source policy.
What this policy enforces
Use this policy when the answer needs a deliberate verification pass before final output instead of a single-pass response.
Evidence model
Bound to the active facts-only policy
This policy does not define admissible evidence on its own.
Failure behavior
Fail closed when verification cannot complete
If verification cannot be completed under the active evidence boundary, the response must fail closed.
Canonical links
Open the linked procedure, prompt file, and adjacent evidence-boundary pages that pair with this verification policy.
Related
Procedure and prompt files for this policy
Open the procedure linked to this policy and any prompt files used with it.
Prompt files (1) Hide prompt files
Non-negotiable rules
These rules define the minimum verification contract before final output.
R1
Evidence requirements come from the active facts-only boundary
Evidence requirements are governed by the active Facts-only evidence boundary.
R2
Verify claims against admissible evidence before output
Claims must be checked against admissible evidence before the final answer is produced.
R3
No simulation
Do not simulate missing evidence, unsupported verification results, or unverifiable completion.
R4
Fail closed under the active boundary
If verification cannot be completed, fail closed according to the exact sentinel or fail-closed behavior required by the active boundary.
Reference
This policy is aligned to the Chain-of-Verification framing as an explicit verification pass before final output.
CoVe (Chain-of-Verification)