feat: Phase G G10 — recalibration rollback plan (Phase G complete)

Makefile

@@ -1,4 +1,4 @@
-.PHONY: help demo test lint ci clean bench-leakage bench-multi-negatives bench-baseline bench-hidden-active bench-cluster-split bench-expert-ablation bench-expert-ablation-500 bench-selectivity bench-feature-decomp bench-gate bench-easy-baseline bench-charge-matched bench-order-dependent bench-precision-at-k bench-per-family bench-simulation-ablation bench-simulation-ablation-within-amp simulation-gate bench-strategy-compare regenerate-all generate phase3 pilot validate-scoring validate-scoring-phase3 validate-scoring-strict external-pilot pilot-confident presynth-qc gold-standard diversity synthesis-order novelty-broad external-consensus questionnaire gate-check ip-report benchmark-card wave0-5-gate-check wave0-5-novelty-audit wave0-5-novelty-audit-v2 wave0-5-panel wave0-5-evidence wave0-5-fill-external wave0-5b-generate wave0-5b-filter recalibration-engine recalibration-engine-dry-run validate-policy-version generate-synthetic-lab-results bump-policy-version generate-review-packet failed-candidate-report safe-publication-filter classify-negative-informativeness validate-rejection-events negative-result-dashboard calibration-audit calibration-audit-example calibration-overfit-check result-quality-filter synthetic-result-policy-check calibration-decision-checklist
+.PHONY: help demo test lint ci clean bench-leakage bench-multi-negatives bench-baseline bench-hidden-active bench-cluster-split bench-expert-ablation bench-expert-ablation-500 bench-selectivity bench-feature-decomp bench-gate bench-easy-baseline bench-charge-matched bench-order-dependent bench-precision-at-k bench-per-family bench-simulation-ablation bench-simulation-ablation-within-amp simulation-gate bench-strategy-compare regenerate-all generate phase3 pilot validate-scoring validate-scoring-phase3 validate-scoring-strict external-pilot pilot-confident presynth-qc gold-standard diversity synthesis-order novelty-broad external-consensus questionnaire gate-check ip-report benchmark-card wave0-5-gate-check wave0-5-novelty-audit wave0-5-novelty-audit-v2 wave0-5-panel wave0-5-evidence wave0-5-fill-external wave0-5b-generate wave0-5b-filter recalibration-engine recalibration-engine-dry-run validate-policy-version generate-synthetic-lab-results bump-policy-version generate-review-packet failed-candidate-report safe-publication-filter classify-negative-informativeness validate-rejection-events negative-result-dashboard calibration-audit calibration-audit-example calibration-overfit-check result-quality-filter synthetic-result-policy-check calibration-decision-checklist calibration-rollback-plan
 
 PYTHON := $(shell [ -f .venv/bin/python ] && echo .venv/bin/python || echo python3)
 PYTEST  := $(shell [ -f .venv/bin/pytest ] && echo .venv/bin/pytest || echo pytest)
@@ -694,12 +694,22 @@ synthetic-result-policy-check:
 	@echo "Synthetic result policy check done. See /tmp/srp_output.json"
 
 calibration-decision-checklist:
-	echo '{"G9-01":true,"G9-02":true,"G9-03":true,"G9-04":true,"G9-05":true,"G9-06":true,"G9-07":true,"G9-08":true,"G9-09":true,"G9-10":true}' > /tmp/checklist_responses.json
+	echo '{"G9-01":true,"G9-02":true,"G9-03":true,"G9-04":true,"G9-05":true,"G9-06":true,"G9-07":true,"G9-08":true,"G9-09":true,"G9-10":true,"G9-11":true}' > /tmp/checklist_responses.json
 	.venv/bin/openamp-foundry calibration-decision-checklist \
 		--checklist-id CHK-2026-001 \
 		--date 2026-07-09 \
 		--reviewer "human-reviewer" \
 		--responses-json /tmp/checklist_responses.json \
 		--out-json /tmp/checklist_output.json \
-		--out-md /tmp/checklist_output.md
+		--out-md /tmp/checklist_output.md; true
 	@echo "Calibration decision checklist done. See /tmp/checklist_output.json"
+
+calibration-rollback-plan:
+	.venv/bin/openamp-foundry calibration-rollback-plan \
+		--plan-id RBK-2026-001 \
+		--version v0.5.87 \
+		--triggered-by RT-01,RT-04 \
+		--notes "AUROC dropped 0.03 after adding new training candidates." \
+		--out-json /tmp/rollback_plan.json \
+		--out-md /tmp/rollback_plan.md
+	@echo "Rollback plan generated. See /tmp/rollback_plan.json"

docs/evidence/METRICS_CURRENT.md

@@ -5,7 +5,8 @@ Machine-readable snapshot: `outputs/metrics_snapshot.json` regenerated with `mak
 > **Purpose:** One authoritative table of current pipeline metrics. If any doc disagrees
 > with this file, this file wins. Updated whenever benchmark/benchmark config changes.
 >
-> **Last updated:** 2026-07-09 (Phase G G9 — calibration decision review checklist — v0.5.87)
+> **Last updated:** 2026-07-09 (Phase G G10 — recalibration rollback plan — v0.5.88)
+> **New in v0.5.88:** Recalibration rollback plan (G10) — `build_rollback_plan()` produces a structured `RollbackPlan` with 5 rollback triggers (RT-01 through RT-05) covering AUROC regression, calibration metric below threshold, safety filter bypass, cohort distribution shift, and unexplained weight delta. Each trigger has id, name, description, threshold, and severity. `RollbackStep` dataclass with step_number, action, responsible, detail, and dry_lab_only. `RollbackPlan` dataclass with plan_id, version, triggered_by, steps, notes, dry_lab_only, and to_dict(). 6 default rollback steps: halt, document, restore, verify, root-cause, and log. CLI (`openamp-foundry calibration-rollback-plan`) accepts `--plan-id`, `--version`, `--triggered-by`, `--notes`, `--out-json`, `--out-md`. JSON + Markdown output. Schema (`schemas/calibration_rollback_plan.schema.json`). `make calibration-rollback-plan` target. 15 tests. 3078 total. This completes Phase G (G1-G10 — calibration and active-learning rigor).
 > **New in v0.5.87:** Calibration decision review checklist (G9) — `build_checklist()` produces a structured `CalibrationDecisionChecklist` with 12 checklist items (10 required) covering data quality, statistical validity, safety consistency, approval, and documentation. Each item has an id, category, question, rationale, and required flag. `CalibrationDecisionChecklist` dataclass tracks responses, notes, overall_pass, and missing_required. CLI (`openamp-foundry calibration-decision-checklist`) accepts `--checklist-id`, `--date`, `--reviewer`, `--responses-json`, `--out-json`, `--out-md`. JSON + Markdown output. Schema (`schemas/calibration_decision_checklist.schema.json`). `make calibration-decision-checklist` target. 14 tests. 3063 total. Makes human review structured and auditable.
 > **New in v0.5.86:** Synthetic-result policy enforcement (G8) — `check_synthetic_result_policy()` and `run_policy_batch()` enforce that synthetic/simulation outputs cannot raise the proof-ladder level of a candidate. Levels 4+ require wet-lab evidence; synthetic/unknown sources are rejected for such proposals. CLI (`openamp-foundry synthetic-result-policy-check`) accepts `--proposals-json`, `--out-json`, `--out-md`. Schema (`schemas/synthetic_result_policy_check.schema.json`). `make synthetic-result-policy-check` target. 27 tests. 3049 total. Anti-overclaim safeguard.
 > **New in v0.5.85:** Result-quality flag propagation (G7) — `assess_result_quality()` and `filter_results_for_calibration()` propagate result-quality flags into the calibration engine. Candidates with contamination or assay interference are excluded; candidates with 2+ quality flags are excluded; candidates with 1 flag are included with caution; clean candidates are fully included. `QUALITY_FLAGS` dictionary (8 flags). `ResultQualityReport` dataclass. CLI (`openamp-foundry result-quality-filter`) accepts `--results-json`, `--out-json`, `--out-md`. JSON + Markdown output. Schema (`schemas/result_quality_report.schema.json`). `make result-quality-filter` target. 27 tests. 3022 total.
@@ -79,7 +80,7 @@ Machine-readable snapshot: `outputs/metrics_snapshot.json` regenerated with `mak
 > **New in v0.5.31:** Added dipeptide-order features for sequence-order awareness. `dipeptide_order_score` achieves AUROC 0.7861 on AMP-vs-scrambled discrimination — the strongest order-dependent feature in the pipeline. Only 7/31 features survive scrambling (amphipathicity/helix-wheel + dipeptide). All composition features are purely position-independent (exactly 0.5000 AUROC on scrambled test).
 > **New in v0.5.30:** Easy baseline benchmark added — charge density alone (AUROC 0.8166) outperforms the full pipeline ensemble (0.7792) on AMP-vs-Swiss-Prot-decoy discrimination. Honest finding documented: expected because pipeline optimizes for safety, not raw discrimination.
 > **New in v0.5.29:** Expanded benchmark to 500 AMPs + 500 composition-matched decoys (n=1000). AUROC 0.7792 (CI₉₅: 0.7505–0.8065) confirms signal generalizes. Cluster-aware CI: 0.746–0.8102. Representative AUROC: 0.778. Standard benchmark (n=191) retained for backward comparison.
-> **Pipeline version:** v0.5.87 (G9)
+> **Pipeline version:** v0.5.88 (G10 — Phase G complete)
 > **Branch:** main
 
 ---

docs/research/ROADMAP.md

@@ -1,5 +1,60 @@
 # Roadmap
 
+## v0.5.88 — Loop 88: Phase G G10 — Recalibration Rollback Plan ✓ (2026-07-09)
+
+`build_rollback_plan()` produces a structured `RollbackPlan` with 5 rollback
+triggers (RT-01 through RT-05) and 6 default rollback steps covering halt,
+document, restore, verify, root-cause, and log. `RollbackPlan` dataclass with
+plan_id, version, triggered_by, steps, notes, dry_lab_only, and to_dict().
+`RollbackStep` dataclass with step_number, action, responsible, detail, and
+dry_lab_only. CLI (`openamp-foundry calibration-rollback-plan`).
+JSON + Markdown output. Schema (`schemas/calibration_rollback_plan.schema.json`).
+`make calibration-rollback-plan` target. 15 tests. 3078 total.
+This completes Phase G (G1-G10 — calibration and active-learning rigor).
+
+Changes:
+- `src/openamp_foundry/calibration/rollback_plan.py` (G10) — Core module with
+  `ROLLBACK_TRIGGERS` list (5 triggers: RT-01 through RT-05), `RollbackStep`
+  dataclass (5 fields: step_number, action, responsible, detail, dry_lab_only),
+  `RollbackPlan` dataclass (6 fields: plan_id, version, triggered_by, steps,
+  notes, dry_lab_only), `DEFAULT_ROLLBACK_STEPS` list (6 steps),
+  `build_rollback_plan()` with trigger ID validation, `write_rollback_plan_json()`
+  and `write_rollback_plan_markdown()` for structured output.
+- `schemas/calibration_rollback_plan.schema.json` (G10) — JSON Schema Draft
+  2020-12 for the rollback plan. Validates all 6 required fields including
+  plan_id, version, triggered_by with RT-NN pattern, steps with ordered
+  actions, and dry_lab_only const=true.
+- `src/openamp_foundry/calibration/__init__.py` — Exports all rollback plan
+  symbols.
+- `src/openamp_foundry/cli/commands/reports.py` — Added
+  `_run_calibration_rollback_plan()` CLI handler with `--plan-id`, `--version`,
+  `--triggered-by`, `--notes`, `--out-json`, `--out-md` flags.
+- `src/openamp_foundry/cli/main.py` — Registered `calibration-rollback-plan`
+  subcommand with all argument flags and dispatch to handler.
+- `Makefile` — Added `calibration-rollback-plan` target with default example
+  writing to `/tmp/rollback_plan.json` and `/tmp/rollback_plan.md`.
+  Added to `.PHONY`.
+- `tests/calibration/test_rollback_plan.py` — 18 tests covering: valid triggers
+  pass, unknown trigger raises ValueError, triggered_by stored correctly, steps
+  include all default steps, extra_steps appended correctly, dry_lab_only always
+  True, ROLLBACK_TRIGGERS count and required fields, plan_id stored correctly,
+  version stored correctly, to_dict has required keys, notes stored correctly,
+  default steps have correct responsible parties, JSON writer, Markdown writer.
+- `docs/evidence/METRICS_CURRENT.md` — v0.5.88 G10 changelog. Test count: 3081.
+- `tests/test_test_count_regression.py` — baseline updated to 3081.
+
+Honest boundaries:
+- The rollback plan restores weight configurations only. It does not undo
+  candidate selections, synthesis decisions, or code-level regressions.
+- Rollback triggers are detection rules, not guarantees. A regression that
+  does not match any trigger may still be harmful.
+- All rollback steps require human review (steps 2, 5, 6 explicitly).
+  Automated rollback without documented human oversight is not permitted.
+- The plan is a procedural framework. Actual rollback execution may require
+  additional context-specific steps.
+- Dry-lab only. Rollback affects computational scoring, not biological
+  activity, safety, or real-world outcomes.
+
 ## v0.5.87 — Loop 87: Phase G G9 — Calibration Decision Review Checklist ✓ (2026-07-09)
 
 `build_checklist()` produces a structured `CalibrationDecisionChecklist` with 12

schemas/calibration_rollback_plan.schema.json

@@ -0,0 +1,77 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "$id": "https://openamp.ai/schemas/calibration_rollback_plan.schema.json",
+  "title": "Calibration Rollback Plan",
+  "description": "Schema for a structured recalibration rollback plan. Documents the triggers, steps, and responsible parties for reverting a recalibration that degrades performance.",
+  "type": "object",
+  "required": [
+    "plan_id",
+    "version",
+    "triggered_by",
+    "steps",
+    "notes",
+    "dry_lab_only"
+  ],
+  "properties": {
+    "plan_id": {
+      "type": "string",
+      "description": "Unique identifier for the rollback plan (e.g. RBK-2026-001)."
+    },
+    "version": {
+      "type": "string",
+      "description": "The calibration version to roll back (e.g. v0.5.87)."
+    },
+    "triggered_by": {
+      "type": "array",
+      "description": "List of rollback trigger IDs that activated this plan.",
+      "items": {
+        "type": "string",
+        "pattern": "^RT-\\d{2}$"
+      },
+      "minItems": 1
+    },
+    "steps": {
+      "type": "array",
+      "description": "Ordered list of rollback steps to execute.",
+      "items": {
+        "type": "object",
+        "required": ["step_number", "action", "responsible", "detail", "dry_lab_only"],
+        "properties": {
+          "step_number": {
+            "type": "integer",
+            "minimum": 1,
+            "description": "Step order in the rollback procedure."
+          },
+          "action": {
+            "type": "string",
+            "description": "Brief action description."
+          },
+          "responsible": {
+            "type": "string",
+            "description": "Who is responsible: human-reviewer, pipeline, or both.",
+            "enum": ["human-reviewer", "pipeline", "both"]
+          },
+          "detail": {
+            "type": "string",
+            "description": "Detailed instructions for executing this step."
+          },
+          "dry_lab_only": {
+            "type": "boolean",
+            "const": true,
+            "description": "All rollback steps are dry-lab operations."
+          }
+        }
+      },
+      "minItems": 1
+    },
+    "notes": {
+      "type": "string",
+      "description": "Free-text notes about the rollback event."
+    },
+    "dry_lab_only": {
+      "type": "boolean",
+      "const": true,
+      "description": "This artifact is computational only and does not constitute biological proof."
+    }
+  }
+}

src/openamp_foundry/calibration/__init__.py

@@ -91,6 +91,16 @@
     write_checklist_markdown,
 )
 
+from openamp_foundry.calibration.rollback_plan import (
+    ROLLBACK_TRIGGERS,
+    DEFAULT_ROLLBACK_STEPS,
+    RollbackStep,
+    RollbackPlan,
+    build_rollback_plan,
+    write_rollback_plan_json,
+    write_rollback_plan_markdown,
+)
+
 __all__ = [
     # Constants
     "MIN_COHORT_SIZE",
@@ -150,4 +160,12 @@
     "build_checklist",
     "write_checklist_json",
     "write_checklist_markdown",
+    # Rollback plan
+    "ROLLBACK_TRIGGERS",
+    "DEFAULT_ROLLBACK_STEPS",
+    "RollbackStep",
+    "RollbackPlan",
+    "build_rollback_plan",
+    "write_rollback_plan_json",
+    "write_rollback_plan_markdown",
 ]