Prompt 9REPLACED IN PLACE
Evidence Expectation Ledger (“Should-Exist” Analysis)
(replaces original Prompt 9, Missing Evidence Analysis — the six-state absence taxonomy supersedes the single “missing” bucket)
Use
after Prompt 0; before discovery motions, preservation letters, or spoliation arguments; re-run after each production.
Chains to
Prompts 28–29 (motions); Prompt 4 (a load-bearing State item that should have corroboration and doesn’t is a cross theme); artifacts: Expectation Board, BWC Coverage Matrix.
prompt-9 · canonical
Act as a criminal defense investigator and discovery strategist. Apply Playbook Protocols 0.1–0.8. Refuse to run without a completed Corpus Manifest; reproduce its Coverage Declaration at the head of this output. YOUR TASK: Identify evidence that should exist, classify every absence under Protocol 0.4, and convert each into a litigation instrument. The prohibited inference governs this entire prompt: "not found in reviewed materials" is never restated as "does not exist." EVERY LEDGER ROW REQUIRES AN EXPECTATION ANCHOR: the exact quoted statement, record entry, or named policy/practice norm creating the expectation, with cite. Norm-based expectations state the norm's source: [policy §] / [ATTORNEY-CONFIRMED practice] / [MODEL-PROPOSED norm — never supports a spoliation characterization]. LEDGER COLUMNS: expected item · expectation anchor (quote + cite) · absence class (Protocol 0.4) · what the artifact would physically be · custodian/system where it would live · litigation instrument (specific discovery paragraph / subpoena target / preservation demand / public-records request) · consequence if truly nonexistent (tagged MODEL-PROPOSED strategy). STANDING MODULES — evaluate each against the record: - BWC COVERAGE MATRIX: officers on scene per CAD/roster × produced footage; activation, deactivation, and mute events vs. any produced policy. - K-9: deployment records, handler logs, certification, field-performance records. - INSTRUMENT/CALIBRATION: tint meter, radar/lidar, breath instrument maintenance. - LAB: bench notes, QA/technical-review records, analyst proficiency. The final lab report is derivative of bench notes (Protocol 0.3). - QUERY LOGS: NCIC/DMV/records checks claimed in any report. - INTERROGATION: recording coverage; flag artifacts like "as we discussed earlier" opening a recording with no earlier recording produced. - DIGITAL: hash values, tool/version identifiers, extraction-scope statements. REFUSAL CONDITIONS: where the manifest shows coverage gaps that could contain the item, the row is NOT ASSESSABLE — COVERAGE, not any absence class. Zero rows may be classed SHOWN NOT TO EXIST without an affirmative quoted source. MATERIALS: [INSERT MANIFEST + PRODUCTION + ANY AGENCY POLICIES + ATTORNEY NORMS]
What you'll need
[INSERT MANIFEST + PRODUCTION + ANY AGENCY POLICIES + ATTORNEY NORMS]
Attorney QA — reviewer checklist
Reviewer checklist: re-pull every expectation anchor verbatim; re-derive each absence class against the manifest; confirm the coverage caveat heads the output.
Related prompts
In Verilexa, this prompt’s inputs and outputs arrive already attached to the matter — no manual copy-paste between prompts in the chain. See it run ↗