Use trimmed transcript for session recap

architecture.md

@@ -18,12 +18,12 @@ Implemented:
 - real Seriatim subprocess adapter
 - real Audita subprocess adapter
 - real Scriptorium subprocess adapter
+- real `trim` stage producing `transcripts/trimmed.json`
 - real `analyze` stage for initial `session_recap` generation
 - optional Scriptorium render diagnostics (`render_debug`) before production run
 
 Still placeholder/future:
 
-- `trim` stage behavior
 - `archive` stage behavior
 - `notify` stage behavior
 - additional Scriptorium artifact types beyond `session_recap`
@@ -162,8 +162,9 @@ Behavior:
 - if `pipeline.scriptorium` is missing, analyze returns a skipped result with metadata
 - if no Scriptorium artifacts are enabled, analyze returns a skipped result with metadata
 - if enabled artifacts exist but `session_recap` is not enabled, analyze fails clearly
-- processed transcript input is resolved from manifest (`polish` output kind `transcript_processed`) when available, otherwise fallback path `work/<session_id>/transcripts/processed.json`
-- processed transcript is validated as JSON with top-level `segments` array
+- `trimmed_transcript` input is resolved from manifest (`trim` output kind `transcript_trimmed`) when available, otherwise fallback path `work/<session_id>/transcripts/trimmed.json`
+- `processed_transcript` input is resolved from manifest (`polish` output kind `transcript_processed`) when available, otherwise fallback path `work/<session_id>/transcripts/processed.json`
+- transcript inputs are validated as JSON with top-level `segments` array
 - configured inputs are resolved by source
 - optional `previous_recap` is omitted when unavailable
 - required `previous_recap` fails before invocation when unavailable