feat(privacy): audited redaction that keeps history's shape and removes its content

Rows in append-only history are never deleted; their content can now be made irrecoverable through one audited path. A redaction replaces the content columns of an event, memory, message, model call, belief or assertion with a fixed marker, keeps identifiers, links, timestamps, scores and breakdowns, writes an append-only ledger row with a SHA-256 of the removed content, appends a redaction event, optionally cascades to derived rows and to the events that quote them, and scrubs stored snapshot archives before re-signing them. noema_redact() is the only write the append-only trigger admits, for UPDATE only, under a transaction-local flag the application never sets; plain updates and all deletes stay refused. Replay labels redacted stimuli instead of inventing them. Cryptographic erasure was considered and deferred with reasons (ADR-0009); docs/privacy.md defines retained, redacted, irrecoverable and deleted and states what backups and earlier exports are not covered. Migration 0023.
2026-08-23 23:52:07 +01:00
parent 9eed6e2a18
commit 7d608c2783
33 changed files with 1245 additions and 45 deletions
@@ -0,0 +1,13 @@
+-- Redacted content cannot be restored; this only removes the machinery.
+DROP FUNCTION IF EXISTS noema_redact(text, uuid, uuid);
+CREATE OR REPLACE FUNCTION noema_forbid_change() RETURNS trigger LANGUAGE plpgsql AS $$
+BEGIN
+  RAISE EXCEPTION 'table % is append-only', TG_TABLE_NAME USING ERRCODE = 'integrity_constraint_violation';
+END $$;
+ALTER TABLE assertions      DROP COLUMN IF EXISTS redaction_id, DROP COLUMN IF EXISTS redacted_at;
+ALTER TABLE beliefs         DROP COLUMN IF EXISTS redaction_id, DROP COLUMN IF EXISTS redacted_at;
+ALTER TABLE llm_calls       DROP COLUMN IF EXISTS redaction_id, DROP COLUMN IF EXISTS redacted_at;
+ALTER TABLE messages        DROP COLUMN IF EXISTS redaction_id, DROP COLUMN IF EXISTS redacted_at;
+ALTER TABLE memories        DROP COLUMN IF EXISTS redaction_id, DROP COLUMN IF EXISTS redacted_at;
+ALTER TABLE cognitive_events DROP COLUMN IF EXISTS redaction_id, DROP COLUMN IF EXISTS redacted_at;
+DROP TABLE IF EXISTS redactions;
@@ -0,0 +1,81 @@
+-- Privacy redaction (review pass §7.2 item 8): the audited maintenance path
+-- ADR-0002 promised. Redaction replaces the sensitive content of a row with a
+-- fixed marker while keeping the row, its identifiers, its links and a hash
+-- of what was removed. It is the only write ever permitted on append-only
+-- tables, it is confined to content columns, and it runs through
+-- noema_redact() so the application cannot reach the bypass by accident.
+
+CREATE TABLE redactions (
+  id           uuid PRIMARY KEY,
+  mind_id      uuid NOT NULL REFERENCES minds(id),
+  target_kind  text NOT NULL CHECK (target_kind IN ('event','memory','message','llm_call','belief','assertion')),
+  target_id    uuid NOT NULL,
+  reason       text NOT NULL,
+  policy       text NOT NULL DEFAULT '',   -- retention or privacy policy invoked
+  content_hash text NOT NULL,              -- sha256 of the removed content, for later attestation
+  requested_by text NOT NULL,
+  cascade_of   uuid REFERENCES redactions(id),
+  created_at   timestamptz NOT NULL DEFAULT now(),
+  UNIQUE (target_kind, target_id)
+);
+CREATE INDEX redactions_mind_idx ON redactions (mind_id, created_at DESC);
+CREATE TRIGGER redactions_append_only BEFORE UPDATE OR DELETE ON redactions
+  FOR EACH ROW EXECUTE FUNCTION noema_forbid_change();
+
+ALTER TABLE cognitive_events ADD COLUMN IF NOT EXISTS redacted_at timestamptz, ADD COLUMN IF NOT EXISTS redaction_id uuid;
+ALTER TABLE memories        ADD COLUMN IF NOT EXISTS redacted_at timestamptz, ADD COLUMN IF NOT EXISTS redaction_id uuid;
+ALTER TABLE messages        ADD COLUMN IF NOT EXISTS redacted_at timestamptz, ADD COLUMN IF NOT EXISTS redaction_id uuid;
+ALTER TABLE llm_calls       ADD COLUMN IF NOT EXISTS redacted_at timestamptz, ADD COLUMN IF NOT EXISTS redaction_id uuid;
+ALTER TABLE beliefs         ADD COLUMN IF NOT EXISTS redacted_at timestamptz, ADD COLUMN IF NOT EXISTS redaction_id uuid;
+ALTER TABLE assertions      ADD COLUMN IF NOT EXISTS redacted_at timestamptz, ADD COLUMN IF NOT EXISTS redaction_id uuid;
+
+-- The append-only trigger admits exactly one kind of UPDATE: a redaction
+-- running under the session flag that only noema_redact() sets. DELETE
+-- stays forbidden in every case.
+CREATE OR REPLACE FUNCTION noema_forbid_change() RETURNS trigger LANGUAGE plpgsql AS $$
+BEGIN
+  IF TG_OP = 'UPDATE' AND current_setting('noema.redaction', true) = 'on' THEN
+    RETURN NEW;
+  END IF;
+  RAISE EXCEPTION 'table % is append-only', TG_TABLE_NAME USING ERRCODE = 'integrity_constraint_violation';
+END $$;
+
+-- noema_redact replaces content columns for one row. It sets the bypass
+-- flag for the current transaction only, touches nothing but content and
+-- the redaction bookkeeping, and returns the number of rows changed.
+CREATE OR REPLACE FUNCTION noema_redact(kind text, target uuid, redaction uuid) RETURNS integer LANGUAGE plpgsql SECURITY DEFINER AS $$
+DECLARE n integer := 0;
+BEGIN
+  PERFORM set_config('noema.redaction', 'on', true);
+  IF kind = 'event' THEN
+    UPDATE cognitive_events SET summary = '[redacted]', payload = '{"redacted":true}'::jsonb,
+      metadata = (SELECT coalesce(jsonb_object_agg(key, value), '{}'::jsonb) FROM jsonb_each(metadata)
+                  WHERE key IN ('conversation_id','message_id','cycle','belief_id','memory_id','integration_id','integration','kind','trust','method','replay_of','replay_id'))
+                 || jsonb_build_object('redacted', true),
+      redacted_at = now(), redaction_id = redaction WHERE id = target AND redacted_at IS NULL;
+  ELSIF kind = 'memory' THEN
+    UPDATE memories SET content = '[redacted]', detail = '{"redacted":true}'::jsonb, redacted_at = now(), redaction_id = redaction WHERE id = target AND redacted_at IS NULL;
+  ELSIF kind = 'message' THEN
+    UPDATE messages SET content = '[redacted]', intent = NULL, redacted_at = now(), redaction_id = redaction WHERE id = target AND redacted_at IS NULL;
+  ELSIF kind = 'llm_call' THEN
+    UPDATE llm_calls SET request = '{"redacted":true}'::jsonb, response = NULL, error = CASE WHEN error = '' THEN '' ELSE '[redacted]' END, redacted_at = now(), redaction_id = redaction WHERE id = target AND redacted_at IS NULL;
+  ELSIF kind = 'belief' THEN
+    UPDATE beliefs SET proposition = '[redacted]', detail = '{"redacted":true}'::jsonb, redacted_at = now(), redaction_id = redaction WHERE id = target AND redacted_at IS NULL;
+    GET DIAGNOSTICS n = ROW_COUNT;
+    IF n > 0 THEN
+      UPDATE belief_versions SET proposition = '[redacted]', reason = '[redacted]' WHERE belief_id = target;
+      UPDATE belief_evidence SET note = '[redacted]' WHERE belief_id = target;
+      UPDATE belief_alternatives SET proposition = '[redacted]', note = '[redacted]' WHERE belief_id = target;
+    END IF;
+    PERFORM set_config('noema.redaction', 'off', true);
+    RETURN n;
+  ELSIF kind = 'assertion' THEN
+    UPDATE assertions SET proposition = '[redacted]', context = '[redacted]', interpretation_notes = '{}', redacted_at = now(), redaction_id = redaction WHERE id = target AND redacted_at IS NULL;
+  ELSE
+    RAISE EXCEPTION 'unknown redaction kind %', kind;
+  END IF;
+  GET DIAGNOSTICS n = ROW_COUNT;
+  PERFORM set_config('noema.redaction', 'off', true);
+  RETURN n;
+END $$;
+REVOKE ALL ON FUNCTION noema_redact(text, uuid, uuid) FROM PUBLIC;