AINA Data Engine Room · Handoff · 2026-06-11

Import Recipes + Semantic Repair Queue Handoff

A runtime-prep layer that turns source inventory and semantic review into executable next queues.

Ali Mehdi Mukadam · co-authored with Codex · branch ali/personalization-engine-mission-2026-06-09

The Single Idea

The engine room now has two more executable runtime-prep surfaces: source import recipes and a semantic repair queue. The previous harvest map said what exists; these artifacts say what to do next with each source and title row.

01 · What Changed

Two New Runtime-Prep Commands

The repo now has source_import_recipes.py, semantic_repair_queue.py, tests, and CLI commands for both. These are local artifact generators, not external import jobs.

Harvest map16 source roots and expected assets.
Recipes16 source-specific import or repair plans.
Semantic gate1,000 sampled title rows.
Repair queuePacket, caveat, display, and source-ref lanes.
02 · Live Artifacts

Ignored Local Outputs

/srv/aina/aina-data-engine-room/artifacts/validation/source_import_recipes_v1.json
/srv/aina/aina-data-engine-room/artifacts/validation/source_import_recipes_v1.jsonl
/srv/aina/aina-data-engine-room/artifacts/validation/semantic_repair_queue_v1.json
/srv/aina/aina-data-engine-room/artifacts/validation/semantic_repair_queue_v1.jsonl
/srv/aina/aina-data-engine-room/artifacts/provenance.jsonl
03 · Import Recipes

Sixteen Source Plans

StateCount
Ready to import now12
Ready for archive diff/provenance mining2
Reference only1
Repair before import1
The one repair-before-import source is jobs_research_source_intelligence, missing project-summary-package/exports/source_intelligence_v1/responsibilities.jsonl.
04 · Repair Queue

The 1,000-Row Sample Is Now Actionable

Queue laneCount
Packet hardening ready247
Deterministic caveat enrichment485
Title display repair233
Source ref repair35

The important shift is that the 268 semantic-repair rows are no longer a single blob. Most are source-backed title/display/context repairs, while only 35 currently need source-reference mining.

05 · Validation

Green Checks

cd /srv/aina/aina-data-engine-room
.venv/bin/python -m ruff check src tests
.venv/bin/python -m pytest -q

Result: All checks passed. and 193 passed.

06 · Next Slices

Where to Continue

  1. Run deterministic title-display repair for the 233 title_display_repair rows.
  2. Run deterministic caveat enrichment for the 485 deterministic_caveat_enrichment rows.
  3. Mine harvest sources for the 35 source_ref_repair rows.
  4. Promote the 247 packet_hardening_ready rows into packet hardening.
  5. Locate or regenerate the missing jobs-research responsibilities export.
cd /srv/aina/aina-data-engine-room
.venv/bin/aina-data-engine --root /srv/aina/aina-data-engine-room source-import-recipes
.venv/bin/aina-data-engine --root /srv/aina/aina-data-engine-room semantic-repair-queue
Where to start

Start with deterministic display repair and caveat enrichment, because together they touch 718 rows without needing another model review first.