conductor(deob_phases): Initialize 3 phase child spec scaffolds

Each child spec is lightweight (~120 lines): references the umbrella, gives the deliverable structure, specifies the inputs/outputs, and the 5-phase pipeline.

Phase 1 (lexicon): refines the warmup's draft into a codified operational spec (lexicon.md + terms_catalog.md + dedup_map.md)
Phase 2 (pilot): applies the lexicon to 2 Pass 1 reports (cs229_building_llms + entropy_epiplexity), captures refinements in pilot_report.md
Phase 3 (apply): applies the refined lexicon to 10 remaining Pass 1 reports + 1 cross-cutting synthesis, final apply_report.md

3-layer deliverable per video: translation (side-by-side) + replacement (re-encoded) + decoder (per-term etymology + form anchor + definition history)
4 verification criteria: lossless, bounded, constructively typed, etymology-cited

conductor/tracks/video_analysis_deob_apply_20260621/spec.md

@@ -0,0 +1,100 @@
+# Track: Video Analysis De-obfuscation — Apply to 10 + Synthesis (2026-06-21)
+
+**Status:** Not started (umbrella + warmup + lexicon + pilot published 2026-06-21)
+**Type:** Research-only child track (Pass 2 of 3, Phase 3 of 3 within Pass 2)
+**Owner:** Tier 2 Tech Lead (execution)
+**Priority:** A (user-blocking; the final phase of Pass 2)
+
+> **Parent:** This is child #3 of the [video_analysis_deob_20260621](../../video_analysis_deob_20260621/) umbrella. See [umbrella spec.md](../../video_analysis_deob_20260621/spec.md) for the full design.
+
+> **Upstream consumer of:** [video_analysis_deob_pilot_20260621](../../video_analysis_deob_pilot_20260621/) (the `pilot_report.md` captures the refined lexicon from the 2-video pilot).
+
+---
+
+## 1. Inputs
+
+| Input | Source |
+|---|---|
+| Refined lexicon (from pilot) | [video_analysis_deob_pilot_20260621/pilot_report.md](../../video_analysis_deob_pilot_20260621/pilot_report.md) |
+| Warmup `prompt_template.md` | [video_analysis_deob_warmup_20260621/prompt_template.md](../../video_analysis_deob_warmup_20260621/prompt_template.md) |
+| Lexicon child deliverables | [video_analysis_deob_lexicon_20260621/lexicon.md](../../video_analysis_deob_lexicon_20260621/lexicon.md) + `terms_catalog.md` + `dedup_map.md` |
+| 10 Pass 1 reports (remaining after pilot) | [video_analysis_<slug>_20260621/report.md](../../) for the 10 slugs not in the pilot |
+| 1 cross-cutting synthesis | [video_analysis_synthesis_20260621/report.md](../../video_analysis_synthesis_20260621/report.md) |
+
+## 2. The 10 remaining slugs (in execution order, per Pass 1)
+
+1. `probability_logic` (A — math foundations)
+2. `score_dynamics_giorgini` (A — math foundations)
+3. `platonic_intelligence_kumar` (B — Platonic AI)
+4. `free_lunches_levin` (B — Platonic AI)
+5. `generic_systems_fields` (C — biological/cognitive)
+6. `brain_counterintuitive` (C — biological/cognitive)
+7. `neural_dynamics_miller` (C — biological/cognitive)
+8. `multiscale_hoffman` (C — biological/cognitive)
+9. `cs336_architectures` (E — Stanford)
+10. `creikey_dl_cv` (D — applied)
+
+Plus the 11th deliverable: the cross-cutting synthesis.
+
+## 3. Deliverables
+
+For each of the 10 videos + 1 synthesis, 3 files in `artifacts/<slug>/`:
+
+| Artifact | Description |
+|---|---|
+| `<slug>_translation.md` | Side-by-side translation table |
+| `<slug>_deobfuscated.md` | Re-encoded report (replacement) |
+| `<slug>_decoder.md` | Per-term decoder |
+
+Plus:
+
+| Artifact | Description |
+|---|---|
+| `apply_report.md` | Final lexicon v2, final process refinements, open questions for Pass 3 |
+
+**Optional (added per child track execution convention):** `plan.md`, `metadata.json`, `state.toml`.
+
+**Total deliverable file count:** 11 × 3 + 1 = 34 files (plus the apply report).
+
+## 4. Pipeline (5 phases)
+
+Per the umbrella spec §5 (Phase 4 of the umbrella). Each phase commits atomically.
+
+- [ ] **Phase 1: Init.** Initialize the child track.
+- [ ] **Phase 2: Apply to A cluster (2 videos: probability_logic, score_dynamics_giorgini).** 2 × 3-layer deliverables. Commit per video.
+- [ ] **Phase 3: Apply to B cluster (2 videos: platonic_intelligence_kumar, free_lunches_levin).** 2 × 3-layer deliverables. Commit per video.
+- [ ] **Phase 4: Apply to C cluster (4 videos: generic_systems_fields, brain_counterintuitive, neural_dynamics_miller, multiscale_hoffman).** 4 × 3-layer deliverables. Commit per video.
+- [ ] **Phase 5: Apply to E + D + synthesis (cs336_architectures, creikey_dl_cv, synthesis).** 3 × 3-layer deliverables. Commit per video/synthesis.
+- [ ] **Phase 6: Refine + user review.** Write `apply_report.md`. User reviews. End-of-track report.
+
+## 5. Per-deliverable structure
+
+Same 8-section structure as the pilot. Same 3-layer format (translation + replacement + decoder). Same 4 verification criteria.
+
+The apply phase uses the **refined lexicon** from the pilot (not the original from the lexicon child). The refinements are in `pilot_report.md`.
+
+## 6. Connections
+
+- **Upstream:** consumes the refined lexicon from the pilot child
+- **Downstream:** Pass 3 (future) consumes the 11 deliverables as input
+
+## 7. Verification (4 criteria per deliverable)
+
+Per the umbrella spec §8 — lossless, bounded, constructively typed, etymology-cited.
+
+Plus:
+- [ ] All 10 videos + 1 synthesis have 3-layer deliverables
+- [ ] `apply_report.md` captures the final lexicon v2
+- [ ] User has reviewed and approved
+- [ ] `state.toml` updated to `status = "completed"`
+
+## 8. Pass 3 handoff
+
+The 11 deliverables + `apply_report.md` are the input to Pass 3 (projection to applied domain). The "lossless preservation" directive carries: Pass 3 has all the bounded, constructive, type-theoretic forms it needs to apply the user's stylistic preferences (handmade/data-oriented/GPGPU + own caveats).
+
+## 9. See also
+
+- [Umbrella spec.md](../../video_analysis_deob_20260621/spec.md)
+- [Umbrella README.md](../../video_analysis_deob_20260621/README.md)
+- [Pilot child spec.md](../../video_analysis_deob_pilot_20260621/spec.md)
+- [Pass 1 campaign spec §0, §11](../../video_analysis_campaign_20260621/spec.md) — multi-pass framing