torusguard 2.1.0 → 2.1.2

package/skills/torusguard/payload/workflows/torusguard-audit.md

@@ -1,55 +1,47 @@
 ---
-description: Static security AST scanning, stable line-shift invariant fingerprinting, root-cause clustering, and 0-100 confidence scoring.
-tools: Read, Grep, Glob, Bash, Write
-version: 1.4.0
+description: Taint-aware static security AST scanning, cross-file interprocedural dataflow, 86+ rules across 22 families, and 7-signal calibrated confidence scoring.
+tools: Read, Grep, Glob, Bash, Write, run_command
+version: 2.0.0
 agent: auditor
 lifecycle-phase: Phase 2 (Static Audit & Clustering)
 required-skills:
   - torusguard-audit
 scripts-binding:
   - .torusguard/scripts/audit_runner.py
-  - .torusguard/scripts/report_sync.py
-  - .torusguard/scripts/run_manager.py
   - .torusguard/scripts/finding_scorer.py
+  - .torusguard/core/cross_file_taint.py
+  - .torusguard/core/confidence.py
+  - .torusguard/core/incremental.py
 ---
 
-# /torusguard audit — Static Security Code Scan & Clustering
+# /torusguard audit — Deep Taint-Aware Static Security Code Scan & Clustering
 
 $ARGUMENTS
 
 ---
 
 ## Objective
-Static security AST scanning, stable fingerprinting, root-cause clustering, throughput telemetry, and 0-100 scoring.
-
-Supported CLI flags:
-- `npx torusguard audit [--target <dir>]` : Specify project root directory
-- `npx torusguard audit --watch` / `-w` : Continuous watch mode, re-scanning on file change
-- `npx torusguard audit --sarif` : Automatically export OASIS SARIF v2.1.0 log
-- `npx torusguard audit --severity <critical|high|medium|low>` : Floor filter for findings
+Execute deep taint-aware static AST analysis across polyglot project files, evaluate code against 86+ canonical security rules across 22 families, trace interprocedural source-to-sink dataflows, assign stable line-shift invariant fingerprints, cluster architectural root causes, synchronize findings with `security_report.md`, and score findings with auditable 7-signal evidence-chain confidence ratings.
 
 ---
 
-## Mandatory Pre-Flight Context Inspection
+## Tri-Mode Execution
 
-Inspect workspace prerequisites before launching static audit:
-1. **Init State (`torusguard.json`):** Assert repository is initialized.
-2. **Active Rules (`rules/active/`):** Confirm rule definitions exist.
-3. **Exclusions:** Assert `node_modules/`, `.venv/`, `dist/`, `.git/` are skipped.
-4. **Run Folder:** Allocate isolated folder in `.torusguard/runs/`.
-5. **Syntax Check:** Check for syntax errors before parsing ASTs.
+| Mode | Command / Tool | Execution Method |
+| :--- | :--- | :--- |
+| **Mode A: Terminal CLI** | `torusguard audit [--incremental] [--watch] [--json]` | Shell execution of TorusGuard audit engine. |
+| **Mode B: AI Chat Slash** | `/torusguard audit` | Conversational guided audit with bounded context inspection. |
+| **Mode C: Native MCP Tool** | `torusguard_audit` | Autonomous agent tool invocation via stdio JSON-RPC. |
 
 ---
 
-## When to Use /torusguard audit
+## Mandatory Pre-Flight Context Inspection
 
-| Trigger Scenario | Recommended Action |
-| :--- | :--- |
-| First-time scan of repository or new branch | Run `/torusguard audit` |
-| Pre-commit review and PR security checks | Run `/torusguard audit` |
-| Uninitialized project | Run `/torusguard init` first |
-| Live endpoint or runtime probing | Run `/torusguard web-validate` |
-| Differential check after patch | Run `/torusguard recheck` |
+Inspect workspace prerequisites before launching static audit:
+1. **Init State (`torusguard.json`):** Assert repository is initialized or run `torusguard init`.
+2. **Active Rules (`rules/`):** Confirm rule definitions exist across the 22 architectural families.
+3. **Exclusions:** Assert `node_modules/`, `.venv/`, `dist/`, `.git/` are skipped.
+4. **Syntax Check:** Check for syntax errors before parsing ASTs.
 
 ---
 
@@ -57,31 +49,19 @@ Inspect workspace prerequisites before launching static audit:
 - Audit discoveries automatically synchronize to `security_report.md` at workspace root.
 - Findings transition into `OPEN 🔴` status with stable invariant fingerprints.
 
-## Execution Steps
-
-1. **Allocate Run Folder:** Run `python .torusguard/scripts/run_manager.py create audit`.
-2. **Scan Codebase ASTs:** Match active rules against source trees.
-3. **Compute Stable Fingerprints:** Hash AST context to produce stable IDs.
-4. **Cluster Root Causes:** Group findings sharing identical sinks.
-5. **Score Confidence (0–100):** Run `python .torusguard/scripts/finding_scorer.py --run <run_dir>`.
-6. **Emit Artifacts:** Write `findings.md`, `findings.json`, and `summary.md`.
-
 ---
 
-## Failure Recovery
-
-- **Zero Rules Active:** Re-run `/torusguard init` to activate rules.
-- **AST Parse Error:** Log syntax error on malformed file, skip, and continue.
-- **Scorer Failure:** Ensure Python 3.10+; verify finding JSON structure.
-- **Halt Trigger:** Abort if run folder cannot be allocated or disk write fails.
-
----
-
-## Hallucination Guard
+## Execution Steps
 
-- ❌ Never invent finding IDs without AST line hashing.
-- ❌ Never flag test fixtures as critical security flaws.
-- ✅ Always calculate scores using `.torusguard/scripts/finding_scorer.py`.
+1. **Launch Audit Scan:**
+   - **Mode A (CLI):** Run `torusguard audit` (or `python .torusguard/scripts/audit_runner.py`). Use `--incremental` for fast diffs.
+   - **Mode B (Chat):** Parse AST sinks and trace sources to sinks using `core.taint_graph` and bounded context windows.
+   - **Mode C (MCP):** Call `torusguard_audit` with `{"target": "."}`.
+2. **Scan Codebase ASTs:** Match 86+ canonical rules across 22 families against source trees.
+3. **Trace Taint Dataflows:** Follow untrusted inputs from sources to dangerous sinks across module boundaries.
+4. **Compute Stable Fingerprints:** Hash AST context to produce stable line-shift invariant IDs (`TG-XXX-<hash12>`).
+5. **Cluster Root Causes:** Group findings into the 22 canonical architectural root causes.
+6. **Synchronize Ground Truth:** Update `security_report.md` at workspace root.
 
 ---
 
@@ -89,12 +69,12 @@ Inspect workspace prerequisites before launching static audit:
 
 ```markdown
 ### 🔎 TorusGuard Static Audit Results
-- **Run ID:** `run-YYYYMMDD-HHMMSS-audit`
-- **Files Scanned:** [Count] source files
+- **Files Scanned:** [Count] source files ([Changed] changed)
+- **Taint Paths Confirmed:** [Count] verified dataflows
 - **Total Findings:** [Count] ([Critical] Critical, [High] High)
 - **Root Cause Clusters:** [Count] architectural issues
-- **Confidence:** [Score]/100
-- **Artifact:** `.torusguard/runs/<run_id>/findings.md`
+- **Confidence:** [Score]/100 (Evidence-Chain Calibrated)
+- **Living Report:** `security_report.md`
 ```
 
 ---
@@ -102,4 +82,4 @@ Inspect workspace prerequisites before launching static audit:
 ## Next Steps
 
 1. Run `/torusguard verify` to validate exploitability and review evidence.
-2. Run `/torusguard harden` to generate minimal surgical remediation plans.
+2. Run `/torusguard harden` to generate minimal surgical remediation plans conforming to the Ponytail protocol (<= 35 additions, <= 25 deletions).

package/skills/torusguard/references/csharp-security.md

@@ -1,41 +1,41 @@
-# TorusGuard Skill Reference: C# & ASP.NET Core Security
-
-> **Loaded When:** A project is identified as a C# / .NET application (`*.csproj`, `*.sln`, ASP.NET Core detected).
-
----
-
-## 🛡️ Key Inspection Areas & Rules
-
-### 1. Authentication & Authorization
-* `TG-AUTH-001`: Ensure controllers and endpoints have `[Authorize]` attributes.
-* Prohibit unvetted `[AllowAnonymous]` on sensitive business logic or user data endpoints.
-
-### 2. Entity Framework Core & SQL Injection
-* `TG-INPUT-002`: Use LINQ expressions or `FromSqlInterpolated` with interpolated string syntax (EF Core converts parameters automatically).
-* Prohibit raw `FromSqlRaw` with un-parameterized string concatenations.
-* `TG-DB-004`: Configure Global Query Filters (`HasQueryFilter(e => e.TenantId == _currentTenant)`) for multi-tenant safety.
-
-### 3. Anti-Forgery & TLS
-* `TG-CSRF-001`: Ensure `[ValidateAntiForgeryToken]` or `[AutoValidateAntiforgeryToken]` is applied to POST/PUT/DELETE actions.
-* Prohibit `DangerousAcceptAnyServerCertificateValidator`.
-
----
-
-## 🛠️ Safe Patterns Summary
-
-```csharp
-// Safe Entity Framework Parameterized Query
-var user = await context.Users
-    .Where(u => u.Id == userId && u.TenantId == currentTenantId)
-    .FirstOrDefaultAsync();
-
-// Safe Global Query Filter in DbContext
-modelBuilder.Entity<Invoice>()
-    .HasQueryFilter(i => i.TenantId == _tenantProvider.GetTenantId());
-
-// Safe Controller Security
-[Authorize]
-[ApiController]
-[Route("api/[controller]")]
-public class InvoicesController : ControllerBase { ... }
-```
+# TorusGuard Skill Reference: C# & ASP.NET Core Security
+
+> **Loaded When:** A project is identified as a C# / .NET application (`*.csproj`, `*.sln`, ASP.NET Core detected).
+
+---
+
+## 🛡️ Key Inspection Areas & Rules
+
+### 1. Authentication & Authorization
+* `TG-AUTH-001`: Ensure controllers and endpoints have `[Authorize]` attributes.
+* Prohibit unvetted `[AllowAnonymous]` on sensitive business logic or user data endpoints.
+
+### 2. Entity Framework Core & SQL Injection
+* `TG-INPUT-002`: Use LINQ expressions or `FromSqlInterpolated` with interpolated string syntax (EF Core converts parameters automatically).
+* Prohibit raw `FromSqlRaw` with un-parameterized string concatenations.
+* `TG-DB-004`: Configure Global Query Filters (`HasQueryFilter(e => e.TenantId == _currentTenant)`) for multi-tenant safety.
+
+### 3. Anti-Forgery & TLS
+* `TG-CSRF-001`: Ensure `[ValidateAntiForgeryToken]` or `[AutoValidateAntiforgeryToken]` is applied to POST/PUT/DELETE actions.
+* Prohibit `DangerousAcceptAnyServerCertificateValidator`.
+
+---
+
+## 🛠️ Safe Patterns Summary
+
+```csharp
+// Safe Entity Framework Parameterized Query
+var user = await context.Users
+    .Where(u => u.Id == userId && u.TenantId == currentTenantId)
+    .FirstOrDefaultAsync();
+
+// Safe Global Query Filter in DbContext
+modelBuilder.Entity<Invoice>()
+    .HasQueryFilter(i => i.TenantId == _tenantProvider.GetTenantId());
+
+// Safe Controller Security
+[Authorize]
+[ApiController]
+[Route("api/[controller]")]
+public class InvoicesController : ControllerBase { ... }
+```

package/skills/torusguard/references/go-security.md

@@ -1,41 +1,41 @@
-# TorusGuard Skill Reference: Go Security
-
-> **Loaded When:** A project is identified as a Go application (`go.mod`, `go.sum`, or Go source files detected).
-
----
-
-## 🛡️ Key Inspection Areas & Rules
-
-### 1. Database & SQL Injection
-* `TG-INPUT-002`: Ensure raw queries use parameterized placeholders (`?`, `$1`), never `fmt.Sprintf` or string concatenation.
-* `TG-DB-004`: Enforce tenant isolation in GORM queries (`db.Where("tenant_id = ?", tenantID)`).
-
-### 2. TLS & Network Security
-* `TG-DIFF-001`: Prohibit `InsecureSkipVerify: true` in `tls.Config`. Always use valid root CA pools.
-* `TG-SSRF-001`: Validate outbound URLs before issuing `http.Get` or `http.Post`.
-
-### 3. Concurrency & Goroutine Safety
-* `TG-BIZ-001`: Ensure context cancellation propagation (`ctx.Done()`) to prevent goroutine leaks.
-* Use mutexes or channels to prevent data races on shared state.
-
-### 4. Input Validation & Path Traversal
-* `TG-INPUT-006`: Sanitize file paths with `filepath.Clean` and check `strings.HasPrefix(cleanPath, baseDir)`.
-
----
-
-## 🛠️ Safe Patterns Summary
-
-```go
-// Safe Parameterized Query
-row := db.QueryRowContext(ctx, "SELECT id, email FROM users WHERE id = ? AND tenant_id = ?", userID, tenantID)
-
-// Safe GORM Scoping
-var invoice Invoice
-result := db.Where("id = ? AND tenant_id = ?", invoiceID, currentTenantID).First(&invoice)
-
-// Safe TLS Client
-tlsConfig := &tls.Config{
-    MinVersion: tls.VersionTLS13,
-}
-client := &http.Client{Transport: &http.Transport{TLSClientConfig: tlsConfig}}
-```
+# TorusGuard Skill Reference: Go Security
+
+> **Loaded When:** A project is identified as a Go application (`go.mod`, `go.sum`, or Go source files detected).
+
+---
+
+## 🛡️ Key Inspection Areas & Rules
+
+### 1. Database & SQL Injection
+* `TG-INPUT-002`: Ensure raw queries use parameterized placeholders (`?`, `$1`), never `fmt.Sprintf` or string concatenation.
+* `TG-DB-004`: Enforce tenant isolation in GORM queries (`db.Where("tenant_id = ?", tenantID)`).
+
+### 2. TLS & Network Security
+* `TG-DIFF-001`: Prohibit `InsecureSkipVerify: true` in `tls.Config`. Always use valid root CA pools.
+* `TG-SSRF-001`: Validate outbound URLs before issuing `http.Get` or `http.Post`.
+
+### 3. Concurrency & Goroutine Safety
+* `TG-BIZ-001`: Ensure context cancellation propagation (`ctx.Done()`) to prevent goroutine leaks.
+* Use mutexes or channels to prevent data races on shared state.
+
+### 4. Input Validation & Path Traversal
+* `TG-INPUT-006`: Sanitize file paths with `filepath.Clean` and check `strings.HasPrefix(cleanPath, baseDir)`.
+
+---
+
+## 🛠️ Safe Patterns Summary
+
+```go
+// Safe Parameterized Query
+row := db.QueryRowContext(ctx, "SELECT id, email FROM users WHERE id = ? AND tenant_id = ?", userID, tenantID)
+
+// Safe GORM Scoping
+var invoice Invoice
+result := db.Where("id = ? AND tenant_id = ?", invoiceID, currentTenantID).First(&invoice)
+
+// Safe TLS Client
+tlsConfig := &tls.Config{
+    MinVersion: tls.VersionTLS13,
+}
+client := &http.Client{Transport: &http.Transport{TLSClientConfig: tlsConfig}}
+```

package/skills/torusguard/references/java-security.md

@@ -1,40 +1,40 @@
-# TorusGuard Skill Reference: Java & Spring Security
-
-> **Loaded When:** A project is identified as a Java/Kotlin application (`pom.xml`, `build.gradle`, Spring Boot detected).
-
----
-
-## 🛡️ Key Inspection Areas & Rules
-
-### 1. Spring Security & CSRF
-* `TG-CSRF-001`: Prohibit `.csrf().disable()` or `AbstractHttpConfigurer::disable` on state-changing browser routes.
-* `TG-PLATFORM-001`: Avoid `@CrossOrigin(origins = "*")` with `allowCredentials = "true"`.
-
-### 2. JPA & Hibernate SQL Injection
-* `TG-INPUT-002`: Use named parameters (`:param`) or `CriteriaBuilder`. Never concatenate strings into JPQL queries.
-* `TG-DB-004`: Enforce tenant isolation via `@Filter` or explicit where clauses.
-
-### 3. Insecure Deserialization & XML
-* Prohibit `ObjectInputStream.readObject()` on untrusted input without look-ahead verification.
-* Disable external entity resolution (`setFeature("http://apache.org/xml/features/disallow-doctype-decl", true)`).
-
----
-
-## 🛠️ Safe Patterns Summary
-
-```java
-// Safe JPA Query
-@Query("SELECT u FROM User u WHERE u.id = :id AND u.tenantId = :tenantId")
-Optional<User> findByIdAndTenantId(@Param("id") Long id, @Param("tenantId") String tenantId);
-
-// Safe Spring Security Configuration
-@Bean
-public SecurityFilterChain filterChain(HttpSecurity http) throws Exception {
-    http
-        .authorizeHttpRequests(auth -> auth
-            .requestMatchers("/public/**").permitAll()
-            .anyRequest().authenticated()
-        );
-    return http.build();
-}
-```
+# TorusGuard Skill Reference: Java & Spring Security
+
+> **Loaded When:** A project is identified as a Java/Kotlin application (`pom.xml`, `build.gradle`, Spring Boot detected).
+
+---
+
+## 🛡️ Key Inspection Areas & Rules
+
+### 1. Spring Security & CSRF
+* `TG-CSRF-001`: Prohibit `.csrf().disable()` or `AbstractHttpConfigurer::disable` on state-changing browser routes.
+* `TG-PLATFORM-001`: Avoid `@CrossOrigin(origins = "*")` with `allowCredentials = "true"`.
+
+### 2. JPA & Hibernate SQL Injection
+* `TG-INPUT-002`: Use named parameters (`:param`) or `CriteriaBuilder`. Never concatenate strings into JPQL queries.
+* `TG-DB-004`: Enforce tenant isolation via `@Filter` or explicit where clauses.
+
+### 3. Insecure Deserialization & XML
+* Prohibit `ObjectInputStream.readObject()` on untrusted input without look-ahead verification.
+* Disable external entity resolution (`setFeature("http://apache.org/xml/features/disallow-doctype-decl", true)`).
+
+---
+
+## 🛠️ Safe Patterns Summary
+
+```java
+// Safe JPA Query
+@Query("SELECT u FROM User u WHERE u.id = :id AND u.tenantId = :tenantId")
+Optional<User> findByIdAndTenantId(@Param("id") Long id, @Param("tenantId") String tenantId);
+
+// Safe Spring Security Configuration
+@Bean
+public SecurityFilterChain filterChain(HttpSecurity http) throws Exception {
+    http
+        .authorizeHttpRequests(auth -> auth
+            .requestMatchers("/public/**").permitAll()
+            .anyRequest().authenticated()
+        );
+    return http.build();
+}
+```

package/skills/torusguard/references/polyglot-security-matrix.md

@@ -1,25 +1,25 @@
-# TorusGuard Universal Polyglot Security Matrix
-
-> **Loaded When:** Any language or framework is detected. Maps universal security invariants across major programming language ecosystems.
-
----
-
-## 🛡️ Cross-Language Invariant Mappings
-
-| Rule Family | Python | TypeScript / Node.js | Go | Rust | Java / Kotlin | C# / .NET |
-|---|---|---|---|---|---|---|
-| **TG-SEC (Secrets)** | `os.environ["KEY"]` | `process.env.KEY` | `os.Getenv("KEY")` | `std::env::var("KEY")` | `System.getenv("KEY")` | `Environment.GetEnvironmentVariable()` |
-| **TG-DB (SQL Injection)** | `%s` / ORM bound | `$1` / Prisma / Drizzle | `?` / `$1` / GORM `?` | `sqlx::query!` / Diesel | Named parameters `:param` | `@p0` / `FromSqlInterpolated` |
-| **TG-DB-004 (Tenant Scope)** | `.filter(tenant=...)` | `where: { tenantId }` | `.Where("tenant_id = ?", tid)` | `filter(tenant_id.eq(tid))` | `WHERE e.tenantId = :tid` | `.Where(x => x.TenantId == tid)` |
-| **TG-AUTH (Access Control)** | `@login_required` / Depends | Middleware auth / Guards | Middleware handlers | Tower layer / extractors | `@PreAuthorize` / Spring Sec | `[Authorize]` attributes |
-| **TG-CSRF (Request Forgery)** | CSRF Middleware | SameSite / csurf | Gorilla CSRF | Anti-CSRF token middleware | Spring Security CSRF | `[ValidateAntiForgeryToken]` |
-| **TG-DIFF (Bypass Comments)** | No `# nosec` | No `// nosec` | No `/* nosec */` | No `#![allow(unsafe)]` | No `@SuppressWarnings` | No pragma disable |
-| **TG-DIFF (TLS Verification)** | No `verify=False` | No `rejectUnauthorized:false` | No `InsecureSkipVerify: true` | No `danger_accept_invalid_certs`| No `TrustAllStrategy` | No `DangerousAcceptAny...` |
-
----
-
-## 🚀 Polyglot Hardening Rules of Thumb
-
-1. **Parameterize Everything:** Never concatenate raw user input into SQL, shell commands, LDAP filters, or template strings, regardless of language.
-2. **Fail Closed on Auth:** Default all endpoints to authenticated; explicitly mark public routes rather than remembering to protect private ones.
-3. **Strict Bounds:** Keep remediation diffs minimal ($\le 35$ additions, $\le 25$ deletions) to ensure auditability across all codebases.
+# TorusGuard Universal Polyglot Security Matrix
+
+> **Loaded When:** Any language or framework is detected. Maps universal security invariants across major programming language ecosystems.
+
+---
+
+## 🛡️ Cross-Language Invariant Mappings
+
+| Rule Family | Python | TypeScript / Node.js | Go | Rust | Java / Kotlin | C# / .NET |
+|---|---|---|---|---|---|---|
+| **TG-SEC (Secrets)** | `os.environ["KEY"]` | `process.env.KEY` | `os.Getenv("KEY")` | `std::env::var("KEY")` | `System.getenv("KEY")` | `Environment.GetEnvironmentVariable()` |
+| **TG-DB (SQL Injection)** | `%s` / ORM bound | `$1` / Prisma / Drizzle | `?` / `$1` / GORM `?` | `sqlx::query!` / Diesel | Named parameters `:param` | `@p0` / `FromSqlInterpolated` |
+| **TG-DB-004 (Tenant Scope)** | `.filter(tenant=...)` | `where: { tenantId }` | `.Where("tenant_id = ?", tid)` | `filter(tenant_id.eq(tid))` | `WHERE e.tenantId = :tid` | `.Where(x => x.TenantId == tid)` |
+| **TG-AUTH (Access Control)** | `@login_required` / Depends | Middleware auth / Guards | Middleware handlers | Tower layer / extractors | `@PreAuthorize` / Spring Sec | `[Authorize]` attributes |
+| **TG-CSRF (Request Forgery)** | CSRF Middleware | SameSite / csurf | Gorilla CSRF | Anti-CSRF token middleware | Spring Security CSRF | `[ValidateAntiForgeryToken]` |
+| **TG-DIFF (Bypass Comments)** | No `# nosec` | No `// nosec` | No `/* nosec */` | No `#![allow(unsafe)]` | No `@SuppressWarnings` | No pragma disable |
+| **TG-DIFF (TLS Verification)** | No `verify=False` | No `rejectUnauthorized:false` | No `InsecureSkipVerify: true` | No `danger_accept_invalid_certs`| No `TrustAllStrategy` | No `DangerousAcceptAny...` |
+
+---
+
+## 🚀 Polyglot Hardening Rules of Thumb
+
+1. **Parameterize Everything:** Never concatenate raw user input into SQL, shell commands, LDAP filters, or template strings, regardless of language.
+2. **Fail Closed on Auth:** Default all endpoints to authenticated; explicitly mark public routes rather than remembering to protect private ones.
+3. **Strict Bounds:** Keep remediation diffs minimal ($\le 35$ additions, $\le 25$ deletions) to ensure auditability across all codebases.

package/skills/torusguard/references/rust-security.md

@@ -1,40 +1,40 @@
-# TorusGuard Skill Reference: Rust Security
-
-> **Loaded When:** A project is identified as a Rust application (`Cargo.toml` or `.rs` files detected).
-
----
-
-## 🛡️ Key Inspection Areas & Rules
-
-### 1. Memory Safety & Unsafe Blocks
-* `TG-INPUT-003`: Audit and restrict `unsafe { ... }` blocks. Must include a `// SAFETY:` rationale comment.
-* Never disable standard borrow checker lints via `#![allow(unsafe_code)]`.
-
-### 2. Database & SQL Queries
-* `TG-INPUT-002`: Use compile-time checked `sqlx::query!` or Diesel query builder methods.
-* `TG-DB-004`: Enforce tenant filtering in queries (`filter(tenant_id.eq(current_tenant_id))`).
-
-### 3. Outbound Requests & TLS
-* `TG-DIFF-001`: Prohibit `danger_accept_invalid_certs(true)` in `reqwest::ClientBuilder`.
-* `TG-SSRF-001`: Enforce IP allowlists before dispatching network calls.
-
----
-
-## 🛠️ Safe Patterns Summary
-
-```rust
-// Safe sqlx Parameterized Query
-let user = sqlx::query_as!(
-    User,
-    "SELECT id, email FROM users WHERE id = $1 AND tenant_id = $2",
-    user_id,
-    current_tenant_id
-)
-.fetch_one(&pool)
-.await?;
-
-// Safe reqwest Client (Strict TLS)
-let client = reqwest::Client::builder()
-    .timeout(std::time::Duration::from_secs(10))
-    .build()?;
-```
+# TorusGuard Skill Reference: Rust Security
+
+> **Loaded When:** A project is identified as a Rust application (`Cargo.toml` or `.rs` files detected).
+
+---
+
+## 🛡️ Key Inspection Areas & Rules
+
+### 1. Memory Safety & Unsafe Blocks
+* `TG-INPUT-003`: Audit and restrict `unsafe { ... }` blocks. Must include a `// SAFETY:` rationale comment.
+* Never disable standard borrow checker lints via `#![allow(unsafe_code)]`.
+
+### 2. Database & SQL Queries
+* `TG-INPUT-002`: Use compile-time checked `sqlx::query!` or Diesel query builder methods.
+* `TG-DB-004`: Enforce tenant filtering in queries (`filter(tenant_id.eq(current_tenant_id))`).
+
+### 3. Outbound Requests & TLS
+* `TG-DIFF-001`: Prohibit `danger_accept_invalid_certs(true)` in `reqwest::ClientBuilder`.
+* `TG-SSRF-001`: Enforce IP allowlists before dispatching network calls.
+
+---
+
+## 🛠️ Safe Patterns Summary
+
+```rust
+// Safe sqlx Parameterized Query
+let user = sqlx::query_as!(
+    User,
+    "SELECT id, email FROM users WHERE id = $1 AND tenant_id = $2",
+    user_id,
+    current_tenant_id
+)
+.fetch_one(&pool)
+.await?;
+
+// Safe reqwest Client (Strict TLS)
+let client = reqwest::Client::builder()
+    .timeout(std::time::Duration::from_secs(10))
+    .build()?;
+```