evidence: add B1 external re-check run template

paper/flagship/third_party_checker_handoff/external_recheck_result_template.json

@@ -0,0 +1,10 @@
+{
+  "runner": "",
+  "date": "",
+  "checker_name": "",
+  "checker_version": "",
+  "inputs_checked": [],
+  "results": [],
+  "notes": "",
+  "status": "template-only"
+}

paper/flagship/third_party_checker_handoff/external_recheck_runbook.md

@@ -0,0 +1,51 @@
+# External Recheck Runbook
+
+## Purpose
+
+This file is a runbook for a future external re-check of the B1 line.
+
+Its purpose is limited: give an external checker implementation or external reviewer a
+minimal execution guide for re-checking the current handoff surface without changing
+canonical B1 or supplementary boundaries.
+
+Canonical B1 remains `1 valid / 3 invalid / 1 demo`.
+
+## Required inputs
+
+Use `checker_input_manifest.json` as the authoritative input list.
+
+- `canonical_b1` defines the minimum canonical inputs that should be checked
+- `supplementary` defines optional supplementary inputs that may be checked without
+  changing canonical B1 counts
+
+## Minimum checks expected
+
+Use `checker_contract.md` as the minimum contract.
+
+At minimum, a future external re-check should be able to cover:
+
+- schema validity
+- unresolved output refs
+- broken evidence-policy refs
+
+## Expected output file shape
+
+The expected output shape is defined by `external_recheck_result_template.json`.
+
+At minimum, the output should record:
+
+- who ran the re-check
+- when it was run
+- which checker name and version were used
+- which inputs were checked
+- what per-input results were returned
+- any notes needed to interpret the run
+- a top-level run status
+
+## Status note
+
+This runbook is for a future external re-check only.
+
+There is no external result included yet.
+
+Status shorthand: no external result is included yet.