diff --git a/SKILL_TEMPLATE.md b/SKILL_TEMPLATE.md
index 8b4f1f4a..f647b5fa 100644
--- a/SKILL_TEMPLATE.md
+++ b/SKILL_TEMPLATE.md
@@ -64,9 +64,10 @@ Hard rules only — falsifiable and enforceable. No "consider" / "may" language.
 ## 4. Remediation
 
 What the agent emits or changes when this fires. Keep complex logic in a
-reference/script file (§7), not inline. If this skill can modify code or
-configuration, classify each remediation path using the repo-level
-`docs/fixer-policy.md` before applying changes.
+reference/script file (§7), not inline. Every fix recommendation must include
+remediation guidance, confidence, blast radius, and behavior-change risk. If this
+skill can modify code or configuration, classify each remediation path using the
+repo-level `docs/fixer-policy.md` before applying changes.
 
 **Before (vulnerable):**
 ```
@@ -78,6 +79,16 @@ configuration, classify each remediation path using the repo-level
 <what the agent should produce>
 ```
 
+**Fix recommendation output:**
+```yaml
+remediations:
+  - guidance: "<concrete remediation steps or patch guidance>"
+    confidence: high                    # low | medium | high
+    blast_radius: "<affected files, users, systems, integrations, or workflows>"
+    behavior_change_risk: medium        # low | medium | high
+    test_strategy: "<optional validation approach>"
+```
+
 ## 5. Verification (falsifiable)
 
 The skill is not "done" until this passes — binary, not aspirational.
@@ -133,6 +144,7 @@ skills/<domain>/<skill-name>/
 - [ ] At least one machine-matchable detection signal (regex / structural)
 - [ ] Rules are hard constraints (no "consider"/"may")
 - [ ] Before/after remediation example present
+- [ ] Every fix recommendation includes `guidance`, `confidence`, `blast_radius`, and `behavior_change_risk`
 - [ ] Falsifiable verification test defined (binary pass/fail)
 - [ ] Gotchas: ≥2 false positives + ≥1 precision trap
 - [ ] `SKILL.md` stays lean; long detail moved to reference files
diff --git a/docs/remediation-output.md b/docs/remediation-output.md
new file mode 100644
index 00000000..aa3a4ce7
--- /dev/null
+++ b/docs/remediation-output.md
@@ -0,0 +1,16 @@
+# Remediation Output Fields
+
+SecuritySkills fix recommendations should be emitted as structured remediation
+items under each finding. Every remediation item must include:
+
+- `guidance`: concrete remediation steps or patch guidance.
+- `confidence`: `low`, `medium`, or `high` confidence that the recommendation is correct for the observed evidence.
+- `blast_radius`: the expected scope of systems, files, users, integrations, data, or workflows affected by the change.
+- `behavior_change_risk`: `low`, `medium`, or `high` risk that the fix changes intended behavior.
+
+`test_strategy` is optional in the normalized schema. Include it when the skill
+can name a concrete validation approach, but keep detailed test strategy policy
+in the dedicated test-strategy workstream.
+
+The machine-readable contract lives in
+[`schemas/finding.schema.json`](../schemas/finding.schema.json).
diff --git a/schemas/finding.schema.json b/schemas/finding.schema.json
new file mode 100644
index 00000000..5822ac6c
--- /dev/null
+++ b/schemas/finding.schema.json
@@ -0,0 +1,114 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "$id": "https://github.com/UnitOneAI/SecuritySkills/schemas/finding.schema.json",
+  "title": "SecuritySkills Normalized Finding Output",
+  "description": "Machine-readable contract for normalized security findings and remediation recommendations emitted by SecuritySkills workflows.",
+  "type": "object",
+  "additionalProperties": false,
+  "required": ["findings"],
+  "properties": {
+    "findings": {
+      "type": "array",
+      "items": {
+        "$ref": "#/$defs/finding"
+      },
+      "description": "Normalized findings produced by a skill run."
+    }
+  },
+  "$defs": {
+    "riskLevel": {
+      "type": "string",
+      "enum": ["low", "medium", "high"]
+    },
+    "nonEmptyString": {
+      "type": "string",
+      "minLength": 1
+    },
+    "finding": {
+      "type": "object",
+      "additionalProperties": false,
+      "required": ["id", "title", "severity", "evidence", "remediations"],
+      "properties": {
+        "id": {
+          "$ref": "#/$defs/nonEmptyString",
+          "description": "Stable finding identifier within the output."
+        },
+        "title": {
+          "$ref": "#/$defs/nonEmptyString"
+        },
+        "severity": {
+          "type": "string",
+          "enum": ["info", "low", "medium", "high", "critical"]
+        },
+        "cwe": {
+          "$ref": "#/$defs/nonEmptyString",
+          "description": "Optional CWE mapping, when applicable."
+        },
+        "framework": {
+          "$ref": "#/$defs/nonEmptyString",
+          "description": "Optional framework/control reference from the skill frontmatter frameworks."
+        },
+        "evidence": {
+          "type": "array",
+          "minItems": 1,
+          "items": {
+            "type": "object",
+            "additionalProperties": false,
+            "required": ["location", "summary"],
+            "properties": {
+              "location": {
+                "$ref": "#/$defs/nonEmptyString"
+              },
+              "summary": {
+                "$ref": "#/$defs/nonEmptyString"
+              }
+            }
+          }
+        },
+        "remediations": {
+          "type": "array",
+          "minItems": 1,
+          "items": {
+            "$ref": "#/$defs/remediation"
+          },
+          "description": "Every fix recommendation for this finding. Each item must include remediation guidance, confidence, blast radius, and behavior-change risk."
+        }
+      },
+      "anyOf": [
+        {
+          "required": ["cwe"]
+        },
+        {
+          "required": ["framework"]
+        }
+      ]
+    },
+    "remediation": {
+      "type": "object",
+      "additionalProperties": false,
+      "required": ["guidance", "confidence", "blast_radius", "behavior_change_risk"],
+      "properties": {
+        "guidance": {
+          "$ref": "#/$defs/nonEmptyString",
+          "description": "Concrete remediation steps or patch guidance for the finding."
+        },
+        "confidence": {
+          "$ref": "#/$defs/riskLevel",
+          "description": "How confident the skill is that this recommendation is correct for the observed evidence."
+        },
+        "blast_radius": {
+          "$ref": "#/$defs/nonEmptyString",
+          "description": "Expected scope of systems, files, users, integrations, data, or workflows affected by the change."
+        },
+        "behavior_change_risk": {
+          "$ref": "#/$defs/riskLevel",
+          "description": "Risk that the remediation changes intended application or operational behavior."
+        },
+        "test_strategy": {
+          "$ref": "#/$defs/nonEmptyString",
+          "description": "Optional validation strategy for the remediation. Kept optional because test strategy requirements are defined separately."
+        }
+      }
+    }
+  }
+}