Refine risk calibration and threat modeling for availability and self-attacks

- Update mantis_calibrate/SKILL.md:
  - Add exceptions to self-attack risk downgrade for lack of non-repudiation and side effects.
  - Simplify repro failure rule to always downgrade to LOW.
  - Add availability_tier enum check for availability-only findings to scale risk multiplier.
  - Output availability_tier in calibration results.
- Update mantis_architecture/SKILL.md to capture component availability requirements in the KB.
- Update mantis_threat_model/SKILL.md to classify availability targets into CRITICAL, STANDARD, or LOW_CRITICALITY tiers in the threat model.
- Update SCHEMA.md to document the availability_tier enum in the Finding Object and define it in the THREAT_MODEL.md structure.

TAG=agy
CONV=af7ca8d9-a795-4f1e-996d-75d7d14ea701

Change-Id: I1602891349abbc7499b05c1a6ac32fcdcff1eda6

Author: nickg-ca

SCHEMA.md

@@ -98,6 +98,9 @@ Fields schema, but with the following specific formatting.*
 -   **`impact_score`** (Integer: 1-5): The calculated impact on CIA triad.
 -   **`likelihood_score`** (Integer: 1-5): The probability of occurrence or
     exploitation.
+-   **`availability_tier`** (Enum, Optional): The availability criticality of
+    the component, if the finding has availability impact.
+    -   Values: `"CRITICAL"`, `"STANDARD"`, `"LOW_CRITICALITY"`, `null`
 -   **`mantis_risk_score`** (Float: 0.1-10.0): The final calculated risk score
     (Hazard).
 -   **`priority`** (String): Qualitative priority bucket (e.g., `"CRITICAL"`,
@@ -145,3 +148,32 @@ trajectory.
 This file serves as an ephemeral inbox/queue for new learnings. It is
 periodically synthesized into the permanent Markdown Knowledge Base
 (`workspace/kb/`) to prevent infinite loops and token bloat.
+
+--------------------------------------------------------------------------------
+
+## 4. Threat Model (`workspace/kb/THREAT_MODEL.md`)
+
+Generated by `/mantis_threat_model`, read by `/mantis_plan` and
+`/mantis_calibrate`.
+
+While this is a Markdown document rather than a strict JSON schema, it serves as
+an inter-stage contract. It must contain the following structured sections to
+ensure downstream agents can accurately parse exposure and criticality:
+
+-   **System Overview Summary:** High-level summary of the system's design and
+    purpose.
+-   **Deployment Intent:** Must state the deployment context (e.g., `Intent:
+    PRODUCTION` or `Intent: SAMPLE_OR_TEST_ONLY`).
+-   **Trust Boundaries:** Definitions of where untrusted inputs meet trusted
+    zones, referencing specific components.
+-   **Threat Actors & Vectors:** Profiles of potential attackers and the
+    boundaries they can access.
+-   **High-Risk Assets & Criticality:** The assets (data, privileges,
+    availability) the attacker wants to compromise.
+    -   **Availability Targets:** For any availability-related assets,
+        explicitly document their **Availability Tier** using one of these
+        values:
+        -   `CRITICAL`: 24/7 immediate operational impact if disrupted.
+        -   `STANDARD`: Important operations; short disruptions are tolerable.
+        -   `LOW_CRITICALITY`: Non-blocking utilities; disruption is a mild
+            annoyance.

mantis_architecture/SKILL.md

@@ -54,11 +54,15 @@ Execute the architecture stage as follows:
         Markdown. Follow these strict paths:
 
         -   `workspace/kb/architecture.md`: High-level data flows, zone
-            definitions, and system design.
+            definitions, system design, and overall availability/uptime
+            requirements (if documented or inferable from configuration like
+            systemd, kubernetes, or load balancers).
         -   `workspace/kb/entities/[component_name].md`: Specific definitions
             for components (e.g., `auth_module.md`). Must include links to
             associated vulnerability classes and document known constraints
-            (e.g., "This module sanitizes input X"). Incorporate trajectory
+            (e.g., "This module sanitizes input X"). Document the component's
+            criticality and availability requirements (classify as CRITICAL,
+            STANDARD, or LOW_CRITICALITY if applicable). Incorporate trajectory
             insights here.
         -   `workspace/kb/vulnerabilities/[CWE-ID_or_BugClass].md`: Descriptions
             of bug classes (e.g., `CWE-79.md` or `Memory-Corruption.md`) that

mantis_calibrate/SKILL.md

@@ -60,6 +60,10 @@ Execute the calibration as follows:
         -   2: Minor loss (e.g., minor information leak, localized disruption).
             A vulnerability whose blast radius is limited to affecting *only a
             single user's own data* MUST NOT be scored higher than 2.
+            *Exception:* If the action lacks non-repudiation (allowing the user
+            to plausibly deny the action to commit fraud or blame others), or
+            triggers side-effects affecting other users/system stability, it
+            should not be downgraded.
         -   1: Negligible impact on CIA, mostly a cosmetic issue. Findings of
             the type "the code is fragile", "lack of defense-in-depth", or
             purely theoretical hygiene issues MUST have an Impact score of 1,
@@ -107,6 +111,12 @@ Execute the calibration as follows:
                     multiplier high.
                 -   If it affects a low-value target (e.g., internal analytics,
                     sandboxed test data), reduce the multiplier (e.g., 0.5).
+                -   **Availability-Specific Context:** If the finding is
+                    availability-only (DoS), check the component's
+                    `availability_tier` in the Threat Model:
+                    -   `LOW_CRITICALITY`: Reduce multiplier to **0.5**.
+                    -   `STANDARD`: Reduce multiplier to **0.8**.
+                    -   `CRITICAL`: Keep multiplier at **1.0**.
                 -   If static/dynamic analysis proves the vulnerable code is
                     effectively "dead code" (never called in runtime execution
                     paths), drastically reduce the multiplier to 0.2.
@@ -143,10 +153,10 @@ Execute the calibration as follows:
     finding's priority to **LOW** (and cap its final score at **2.0**) if it
     meets any of the following "weak finding" criteria:
 
-    -   **Speculative Viability on Repro Failure:** The reproduction failed
-        (`repro_status: "failed_to_reproduce"`), and the reasoning for why it is
-        still viable in production is highly speculative, theoretical, or relies
-        on unverified assumptions about downstream systems.
+    -   **Reproduction Failure:** The reproduction failed (`repro_status:
+        "failed_to_reproduce"`). If the agent cannot successfully reproduce the
+        vulnerability, it MUST be force-downgraded regardless of any theoretical
+        arguments for production viability.
     -   **Minor Configuration Hygiene:** The issue represents a minor deviation
         from best-practice configuration (e.g., slightly loose permissions on an
         internal directory, lack of modern encryption on low-value internal
@@ -176,7 +186,8 @@ Execute the calibration as follows:
     -   **LOW (0.1 - 2.9):** Low priority. Minimal hazard. **Any finding of the
         type "the code is fragile", purely hygiene/defense-in-depth, or one that
         exclusively affects a single user's own data MUST be capped at LOW
-        priority regardless of the calculated score.**
+        priority regardless of the calculated score (unless the exception for
+        lack of non-repudiation or broader side-effects applies).**
 
 5.  **Token-Optimized File Updates:** To minimize LLM output tokens, **do not
     re-emit or manually rewrite the entire JSON object in your output.**
@@ -191,6 +202,7 @@ Execute the calibration as follows:
 
     -   `"impact_score"` (1-5)
     -   `"likelihood_score"` (1-5)
+    -   `"availability_tier"` (CRITICAL, STANDARD, LOW_CRITICALITY or null)
     -   `"mantis_risk_score"` (the final Hazard score)
     -   `"priority"` (CRITICAL, HIGH, MEDIUM, LOW)
     -   `"outrage_commentary"` (your reasoning about the outrage factor)

mantis_threat_model/SKILL.md

@@ -66,7 +66,12 @@ Execute the threat modeling process as follows:
         (e.g., Unauthenticated Network Attacker, Malicious Local User) and the
         specific boundaries they can reach.
     -   **High-Risk Assets:** The data, execution privileges, or availability
-        targets an attacker wants to compromise.
+        targets an attacker wants to compromise. **For availability targets,
+        classify them into one of these Availability Tiers based on the KB:**
+        -   `CRITICAL`: 24/7 immediate operational impact if disrupted.
+        -   `STANDARD`: Important operations; short downtime is tolerable.
+        -   `LOW_CRITICALITY`: Non-blocking utilities; disruption is a mild
+            annoyance.
 
 Save your final output directly to `workspace/kb/THREAT_MODEL.md`. When
 complete, notify the user.