Three more environments, and probe.py to gate all four

canary-trap, fault-localisation and schema-migration, in the three domains the
Environments Hub has nothing in: contamination, operations, data engineering.
None is a port. Each is graded on a slice the model never read.

canary-trap runs a probe set against a model that memorised the corpus AND one
that learned the same facts from a reworded copy with every identifier
regenerated. The obvious probe -- plant a GUID, ask for it back -- is sound and
caps at 0.525 because it cannot see the paraphrase. It scored higher than that
at first: facts with no distinct rewording had the paraphrase restating the
corpus verbatim, so GUID probes caught it by accident. Every fact now carries a
"core" token present in both wordings and in no public one, asserted at import,
which is also what makes the ceiling reachable.

fault-localisation puts the fault upstream and the noise downstream, the way
incidents actually present. The broken service logs one line and goes quiet
while its dependents log a dozen timeouts, so ranking by error volume answers
the victim -- 0/64 incidents have the loudest service as the root, and probe.py
asserts it rather than trusting the generator to stay that way.

schema-migration executes the model's SQL against forty held-out rows carrying
thousands separators, negatives, a unit with a slash in it and a value with no
unit. Row preservation started as a reward beside fidelity and paid 0.15 for
adding two empty columns and touching nothing; it is a multiplier now, so
losing rows still zeroes a perfect migration and keeping them earns nothing on
its own.

probe.py is house rule 3 as an exit code: 0.000 for inaction and 1.000 for an
oracle, or it returns 1. It loads the leaf modules through a namespace shim
rather than the packages, so it runs without verifiers installed -- a self-check
that needs the training stack is a self-check nobody runs.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

environments/redaction_pressure/redaction_pressure/scan.py

@@ -23,6 +23,9 @@ otherwise hang a rollout until the harness killed it.
 
 from __future__ import annotations
 
+import json
+import re
+
 from dataclasses import dataclass
 
 import regex
@@ -134,3 +137,24 @@ def measure(records: list[Record], rules: list[Rule]) -> Outcome:
         total += len(innocent)
         lost += sum(1 for w in innocent if w not in redacted)
     return Outcome(residual, secrets, collateral, decoys, lost, total, invalid)
+
+
+_BLOCK = re.compile(r"```(?:json)?\s*\n(.*?)```", re.DOTALL)
+
+
+def parse_rules(reply: str) -> list[Rule]:
+    """The last JSON array in the reply. A reply with no parsable ruleset is an empty
+    ruleset, not an error — it scores what doing nothing scores."""
+    candidates = _BLOCK.findall(reply or "")
+    raw = candidates[-1] if candidates else (reply or "")
+    try:
+        parsed = json.loads(raw.strip())
+    except json.JSONDecodeError:
+        return []
+    if not isinstance(parsed, list):
+        return []
+    rules = []
+    for item in parsed:
+        if isinstance(item, dict) and isinstance(item.get("pattern"), str):
+            rules.append(Rule(item["pattern"], str(item.get("replacement", ""))))
+    return rules

environments/redaction_pressure/redaction_pressure/taskset.py

@@ -21,7 +21,6 @@ and grades in milliseconds. Deterministic from the task seed.
 
 from __future__ import annotations
 
-import json
 import re
 from typing import ClassVar
 
@@ -30,7 +29,7 @@ from pydantic import Field
 import verifiers.v1 as vf
 
 from redaction_pressure.corpus import build_slices
-from redaction_pressure.scan import MAX_RULES, Rule, measure
+from redaction_pressure.scan import MAX_RULES, Rule, measure, parse_rules
 
 SYSTEM = f"""You are writing redaction rules for a support-ticket corpus.
 
@@ -48,27 +47,6 @@ Example shape:
 [{{"pattern": "\\\\bfoo-\\\\d+\\\\b", "replacement": "[REF]"}}]
 ```"""
 
-_BLOCK = re.compile(r"```(?:json)?\s*\n(.*?)```", re.DOTALL)
-
-
-def parse_rules(reply: str) -> list[Rule]:
-    """The last JSON array in the reply. A reply with no parsable ruleset is an empty
-    ruleset, not an error — it scores what doing nothing scores."""
-    candidates = _BLOCK.findall(reply or "")
-    raw = candidates[-1] if candidates else (reply or "")
-    try:
-        parsed = json.loads(raw.strip())
-    except json.JSONDecodeError:
-        return []
-    if not isinstance(parsed, list):
-        return []
-    rules = []
-    for item in parsed:
-        if isinstance(item, dict) and isinstance(item.get("pattern"), str):
-            rules.append(Rule(item["pattern"], str(item.get("replacement", ""))))
-    return rules
-
-
 class RedactionData(vf.TaskData):
     seed: int
     """Rebuilds both slices exactly; the held-out records are never serialized here."""