torusguard 2.1.0 → 2.1.2

package/skills/torusguard/payload/skills/torusguard-redos/SKILL.md

@@ -1,91 +1,91 @@
----
-name: torusguard-redos
-description: Analyzes regular expressions across JavaScript, TypeScript, Python, and Go for catastrophic exponential backtracking and ReDoS vulnerabilities via CLI, Chat, or MCP.
-version: 2.0.0
-workflow: .torusguard/workflows/redos.md
-tools: Read, Grep, Glob, Write, run_command
-scripts-binding:
-  - internal/scanner/redos.go
-  - cmd/torusguard/main.go
-  - cmd/torusguard/mcp.go
----
-
-# TorusGuard ReDoS Complexity Scanner
-
-## Objective
-Detect and remediate catastrophic exponential ($O(2^n)$) and polynomial ($O(n^k)$) backtracking regular expressions in web applications. Prevents thread starvation, CPU exhaustion, and availability denial of service attacks caused by untrusted user input matching evil regex patterns.
-
----
-
-## Tri-Mode Execution
-
-### Mode A: Automated CLI Execution
-Run ReDoS scanning against source code files:
-```bash
-# Scan current workspace for catastrophic regex backtracking
-torusguard redos
-
-# Scan specific directory or source tree
-torusguard redos --target ./src
-```
-
-### Mode B: In-Session AI Chat Slash Command
-Run `/torusguard redos` in chat.
-The agent executes the compiled Go ReDoS analyzer or MCP tool to locate dangerous nested quantifiers, overlapping alternations, and unanchored expressions.
-
-### Mode C: Native MCP Tool Call
-MCP-enabled coding agents (Antigravity, Cursor, Windsurf, Claude Code) call:
-```json
-{
-  "tool": "torusguard_redos",
-  "arguments": {
-    "target": "."
-  }
-}
-```
-
----
-
-## Supported Patterns & Invariants
-- **TG-REDOS-001 (Nested Quantifier Catastrophic Backtracking):** Detects nested repetitions such as `(a+)+`, `([a-zA-Z0-9]+)*`, or `((foo)*)+` that cause exponential evaluation time $O(2^n)$ when matching non-matching suffixes (e.g. `aaaaaaaaaaaaaaaa!`).
-- **TG-REDOS-002 (Overlapping Alternation with Outer Repetition):** Detects patterns like `(a|ab)+` or `(user|username)*` where multiple branches match identical prefixes, causing exponential branch exploration.
-
----
-
-## 🏛️ OpenCodeReview Hybrid Architecture Integration
-- **First-Principles NFA/DFA Complexity Analysis:** Scans regex literals and constructor invocations across JS, TS, Python, and Go.
-- **Engine-Aware Triage:** Flags backtracking engines (V8, Node.js, Python `re`, PCRE) as high risk while noting linear-time DFA engines (Go `regexp`/RE2, Rust `regex`).
-- **Surgical Patching:** Replaces evil regexes with bounded, non-overlapping expressions or length pre-checks (≤35 additions, ≤25 deletions).
-
----
-
-## 🚨 LLM Trap Table
-
-| Pattern | What AI Does Wrong | What Is Actually Correct |
-| :--- | :--- | :--- |
-| **Nested Quantifiers** | Writes `([a-zA-Z0-9_]+)*` or `(https?://.+)*`, causing $O(2^n)$ catastrophic backtracking on malicious input. | Flatten expressions: `[a-zA-Z0-9_]*` or use non-overlapping delimiters like `[^/\s]+`. |
-| **Overlapping Alternation** | Combines overlapping options inside repetition: `(a\|aa)+` or `(\w+\|\d+)+`. | Disjoint branch design: ensure alternatives cannot consume the same character prefix. |
-| **Missing Input Length Guards** | Evaluates complex regexes against arbitrarily long input strings from untrusted HTTP bodies. | Impose input bounds (`if (input.length > 256) return false`) before regex evaluation. |
-| **Unanchored Wildcards** | Writes `.*foo.*` without anchors, leading to quadratic scan across long documents. | Anchor patterns with `^` and `$` whenever full string matching is intended. |
-| **Engine Indifference** | Assumes all regex engines backtrack identically; doesn't know Go RE2 is linear $O(n)$ while Node V8 is vulnerable. | Tailor remediation to target runtime: use atomic groups/lookahead in V8/Python or switch to parser combinators. |
-
----
-
-## ✅ Pre-Flight Self-Audit
-
-Before concluding a ReDoS complexity audit, verify:
-- [ ] Were all regex literals and `RegExp` / `re.compile` declarations inspected?
-- [ ] Are any nested quantifiers (`(x+)+`, `(x*)*`) present in the codebase?
-- [ ] Are alternatives in grouped alternations strictly mutually exclusive?
-- [ ] Is untrusted input strictly bounded in length before regex execution?
-- [ ] Are remediations validated within Ponytail diff bounds?
-
----
-
-## 🔁 VBC Protocol (Verify → Build → Confirm)
-
-```
-VERIFY: Identify regex patterns evaluated on untrusted user inputs.
-BUILD:  Execute torusguard redos or torusguard_redos to detect exponential backtracking structures.
-CONFIRM: Refactor pattern into deterministic, non-overlapping tokens, adding length guards and testing with evil payloads.
-```
+---
+name: torusguard-redos
+description: Analyzes regular expressions across JavaScript, TypeScript, Python, and Go for catastrophic exponential backtracking and ReDoS vulnerabilities via CLI, Chat, or MCP.
+version: 2.0.0
+workflow: .torusguard/workflows/redos.md
+tools: Read, Grep, Glob, Write, run_command
+scripts-binding:
+  - internal/scanner/redos.go
+  - cmd/torusguard/main.go
+  - cmd/torusguard/mcp.go
+---
+
+# TorusGuard ReDoS Complexity Scanner
+
+## Objective
+Detect and remediate catastrophic exponential ($O(2^n)$) and polynomial ($O(n^k)$) backtracking regular expressions in web applications. Prevents thread starvation, CPU exhaustion, and availability denial of service attacks caused by untrusted user input matching evil regex patterns.
+
+---
+
+## Tri-Mode Execution
+
+### Mode A: Automated CLI Execution
+Run ReDoS scanning against source code files:
+```bash
+# Scan current workspace for catastrophic regex backtracking
+torusguard redos
+
+# Scan specific directory or source tree
+torusguard redos --target ./src
+```
+
+### Mode B: In-Session AI Chat Slash Command
+Run `/torusguard redos` in chat.
+The agent executes the compiled Go ReDoS analyzer or MCP tool to locate dangerous nested quantifiers, overlapping alternations, and unanchored expressions.
+
+### Mode C: Native MCP Tool Call
+MCP-enabled coding agents (Antigravity, Cursor, Windsurf, Claude Code) call:
+```json
+{
+  "tool": "torusguard_redos",
+  "arguments": {
+    "target": "."
+  }
+}
+```
+
+---
+
+## Supported Patterns & Invariants
+- **TG-REDOS-001 (Nested Quantifier Catastrophic Backtracking):** Detects nested repetitions such as `(a+)+`, `([a-zA-Z0-9]+)*`, or `((foo)*)+` that cause exponential evaluation time $O(2^n)$ when matching non-matching suffixes (e.g. `aaaaaaaaaaaaaaaa!`).
+- **TG-REDOS-002 (Overlapping Alternation with Outer Repetition):** Detects patterns like `(a|ab)+` or `(user|username)*` where multiple branches match identical prefixes, causing exponential branch exploration.
+
+---
+
+## 🏛️ OpenCodeReview Hybrid Architecture Integration
+- **First-Principles NFA/DFA Complexity Analysis:** Scans regex literals and constructor invocations across JS, TS, Python, and Go.
+- **Engine-Aware Triage:** Flags backtracking engines (V8, Node.js, Python `re`, PCRE) as high risk while noting linear-time DFA engines (Go `regexp`/RE2, Rust `regex`).
+- **Surgical Patching:** Replaces evil regexes with bounded, non-overlapping expressions or length pre-checks (≤35 additions, ≤25 deletions).
+
+---
+
+## 🚨 LLM Trap Table
+
+| Pattern | What AI Does Wrong | What Is Actually Correct |
+| :--- | :--- | :--- |
+| **Nested Quantifiers** | Writes `([a-zA-Z0-9_]+)*` or `(https?://.+)*`, causing $O(2^n)$ catastrophic backtracking on malicious input. | Flatten expressions: `[a-zA-Z0-9_]*` or use non-overlapping delimiters like `[^/\s]+`. |
+| **Overlapping Alternation** | Combines overlapping options inside repetition: `(a\|aa)+` or `(\w+\|\d+)+`. | Disjoint branch design: ensure alternatives cannot consume the same character prefix. |
+| **Missing Input Length Guards** | Evaluates complex regexes against arbitrarily long input strings from untrusted HTTP bodies. | Impose input bounds (`if (input.length > 256) return false`) before regex evaluation. |
+| **Unanchored Wildcards** | Writes `.*foo.*` without anchors, leading to quadratic scan across long documents. | Anchor patterns with `^` and `$` whenever full string matching is intended. |
+| **Engine Indifference** | Assumes all regex engines backtrack identically; doesn't know Go RE2 is linear $O(n)$ while Node V8 is vulnerable. | Tailor remediation to target runtime: use atomic groups/lookahead in V8/Python or switch to parser combinators. |
+
+---
+
+## ✅ Pre-Flight Self-Audit
+
+Before concluding a ReDoS complexity audit, verify:
+- [ ] Were all regex literals and `RegExp` / `re.compile` declarations inspected?
+- [ ] Are any nested quantifiers (`(x+)+`, `(x*)*`) present in the codebase?
+- [ ] Are alternatives in grouped alternations strictly mutually exclusive?
+- [ ] Is untrusted input strictly bounded in length before regex execution?
+- [ ] Are remediations validated within Ponytail diff bounds?
+
+---
+
+## 🔁 VBC Protocol (Verify → Build → Confirm)
+
+```
+VERIFY: Identify regex patterns evaluated on untrusted user inputs.
+BUILD:  Execute torusguard redos or torusguard_redos to detect exponential backtracking structures.
+CONFIRM: Refactor pattern into deterministic, non-overlapping tokens, adding length guards and testing with evil payloads.
+```

package/skills/torusguard/payload/skills/torusguard-review/SKILL.md

@@ -0,0 +1,27 @@
+---
+name: torusguard-review
+description: Differential PR and Git diff incremental security review — detects newly introduced vulnerabilities between Git branches or commits with net security score deltas.
+tools: Read, Grep, Glob, Bash, Edit, Write
+version: 2.1.2
+last-updated: 2026-09-30
+skills:
+  - torusguard
+  - torusguard-audit
+---
+
+# TorusGuard Review — Differential Incremental PR Review
+
+TorusGuard Review performs rapid, surgical security review on Git changes (e.g. against `HEAD~1`, `main`, or PR target branch), analyzing only modified lines to prevent introducing security regressions.
+
+## Invocation
+
+- **Mode A (Terminal CLI):** `torusguard review [--diff <ref>]`
+- **Mode B (AI Chat Slash Command):** `/torusguard review` or `/torusguard-review`
+- **Mode C (Native MCP Tool):** `torusguard_review(target=".", diff_ref="HEAD~1")`
+
+## Core Capabilities
+
+1. **Sub-300ms Turnaround:** Only parses and checks modified/added lines in Git diffs.
+2. **Net Security Score Delta:** Tracks introduced vs. resolved vulnerabilities (`+0 Introduced, -2 Resolved`).
+3. **PR Gate Decision:** Blocks CI/CD pipelines if Critical or High severity flaws are introduced.
+4. **Alibaba OpenCodeReview Parity:** Outputs ready-to-post pull request inline review comments.

package/skills/torusguard/payload/skills/torusguard-threatmodel/SKILL.md

@@ -0,0 +1,27 @@
+---
+name: torusguard-threatmodel
+description: Synthesizes an architectural STRIDE threat model, interactive Mermaid Data Flow Diagram (DFD), and trust boundary register into SECURITY_THREAT_MODEL.md.
+tools: Read, Grep, Glob, Bash, Edit, Write
+version: 2.1.2
+last-updated: 2026-09-30
+skills:
+  - torusguard
+  - torusguard-audit
+---
+
+# TorusGuard Threat Model — STRIDE & DFD Synthesizer
+
+TorusGuard Threat Model automatically maps routes, trust boundaries, datastores, and egress sinks to construct an architectural Data Flow Diagram and formal STRIDE threat model.
+
+## Invocation
+
+- **Mode A (Terminal CLI):** `torusguard threatmodel`
+- **Mode B (AI Chat Slash Command):** `/torusguard threatmodel` or `/torusguard-threatmodel`
+- **Mode C (Native MCP Tool):** `torusguard_threatmodel(target=".")`
+
+## Core Capabilities
+
+1. **Automated DFD Generation:** Renders interactive Mermaid diagram showing Zones 1–4 (Clients, Gateways, Core Services, Datastores).
+2. **STRIDE Risk Ledger:** Evaluates threats across Spoofing, Tampering, Repudiation, Information Disclosure, Denial of Service, and Elevation of Privilege.
+3. **Multi-Modal Vision Integration:** Discovers architectural diagrams and verifies that visual trust boundaries match code execution paths.
+4. **Living Compliance Artifact:** Emits `SECURITY_THREAT_MODEL.md` at workspace root.

package/skills/torusguard/payload/workflows/ai-guard.md

@@ -1,31 +1,31 @@
-# /torusguard-ai-guard — AI Application & RAG Pipeline Guard
-
-$ARGUMENTS
-
----
-
-## Objective
-Detect and remediate prompt injection, unsandboxed autonomous tool executions, and cross-tenant vector contamination across AI agent and RAG architectures.
-
----
-
-## Tri-Mode Parity
-- **Mode A (Terminal CLI):** `torusguard ai-guard [target]`
-- **Mode B (AI Chat Slash Command):** `/torusguard ai-guard [target]`
-- **Mode C (Native MCP Tool):** `torusguard_ai_guard`
-
----
-
-## Execution Steps
-
-1. **Discover LLM & Agent Call Sites:** Locate AI framework usage (OpenAI, Anthropic, LangChain, LlamaIndex, Vercel AI SDK).
-2. **Audit Prompt Templates:**
-   - Detect raw user string concatenation into system instructions (`TG-AGENT-001`).
-   - Flag raw document chunk ingestion in RAG without inert delimiters (`TG-RAG-002`).
-3. **Audit Tool Registry & Dispatchers:**
-   - Check for unsandboxed shell, file, or database tool calls lacking Human Gates (`TG-AGENT-002`).
-   - Enforce Zod/Pydantic schema validation on all tool argument payloads (`TG-AGENT-003`).
-4. **Audit Vector Store Queries:**
-   - Assert all similarity searches (`pgvector`, `pinecone`, `qdrant`, `chroma`) include tenant/user metadata filters (`TG-RAG-001`).
-   - Flag ingestion of unvalidated external document payloads (`TG-RAG-003`).
-5. **Formulate Patches & Synchronize:** Propose surgical fixes and record findings in `security_report.md`.
+# /torusguard-ai-guard — AI Application & RAG Pipeline Guard
+
+$ARGUMENTS
+
+---
+
+## Objective
+Detect and remediate prompt injection, unsandboxed autonomous tool executions, and cross-tenant vector contamination across AI agent and RAG architectures.
+
+---
+
+## Tri-Mode Parity
+- **Mode A (Terminal CLI):** `torusguard ai-guard [target]`
+- **Mode B (AI Chat Slash Command):** `/torusguard ai-guard [target]`
+- **Mode C (Native MCP Tool):** `torusguard_ai_guard`
+
+---
+
+## Execution Steps
+
+1. **Discover LLM & Agent Call Sites:** Locate AI framework usage (OpenAI, Anthropic, LangChain, LlamaIndex, Vercel AI SDK).
+2. **Audit Prompt Templates:**
+   - Detect raw user string concatenation into system instructions (`TG-AGENT-001`).
+   - Flag raw document chunk ingestion in RAG without inert delimiters (`TG-RAG-002`).
+3. **Audit Tool Registry & Dispatchers:**
+   - Check for unsandboxed shell, file, or database tool calls lacking Human Gates (`TG-AGENT-002`).
+   - Enforce Zod/Pydantic schema validation on all tool argument payloads (`TG-AGENT-003`).
+4. **Audit Vector Store Queries:**
+   - Assert all similarity searches (`pgvector`, `pinecone`, `qdrant`, `chroma`) include tenant/user metadata filters (`TG-RAG-001`).
+   - Flag ingestion of unvalidated external document payloads (`TG-RAG-003`).
+5. **Formulate Patches & Synchronize:** Propose surgical fixes and record findings in `security_report.md`.

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

@@ -1,5 +1,5 @@
 ---
-description: Static security AST scanning, stable line-shift invariant fingerprinting, root-cause clustering, and 0-100 confidence scoring.
+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
@@ -7,19 +7,21 @@ lifecycle-phase: Phase 2 (Static Audit & Clustering)
 required-skills:
   - torusguard-audit
 scripts-binding:
-  - internal/scanner/scanner.go
-  - cmd/torusguard/main.go
-  - cmd/torusguard/mcp.go
+  - .torusguard/scripts/audit_runner.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
-Execute static AST analysis across polyglot project files, evaluate code against 74 canonical security rules across 18 families, assign stable line-shift invariant fingerprints, cluster architectural root causes, synchronize findings with `security_report.md`, and score findings with auditable 0–100 confidence ratings.
+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.
 
 ---
 
@@ -27,7 +29,7 @@ Execute static AST analysis across polyglot project files, evaluate code against
 
 | Mode | Command / Tool | Execution Method |
 | :--- | :--- | :--- |
-| **Mode A: Terminal CLI** | `torusguard audit [--include-tests] [--json]` | Shell execution of compiled Go binary. |
+| **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. |
 
@@ -37,7 +39,7 @@ Execute static AST analysis across polyglot project files, evaluate code against
 
 Inspect workspace prerequisites before launching static audit:
 1. **Init State (`torusguard.json`):** Assert repository is initialized or run `torusguard init`.
-2. **Active Rules (`rules/active/`):** Confirm rule definitions exist.
+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.
 
@@ -52,13 +54,14 @@ Inspect workspace prerequisites before launching static audit:
 ## Execution Steps
 
 1. **Launch Audit Scan:**
-   - **Mode A (CLI):** Run `torusguard audit` in terminal.
-   - **Mode B (Chat):** Parse AST sinks using `grep_search` and bounded `ExtractContext`.
+   - **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 74 canonical rules across 18 families against source trees.
-3. **Compute Stable Fingerprints:** Hash AST context to produce stable line-shift invariant IDs.
-4. **Cluster Root Causes:** Group findings sharing identical sinks or causal architecture.
-5. **Synchronize Ground Truth:** Update `security_report.md` at workspace root.
+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.
 
 ---
 
@@ -66,10 +69,11 @@ Inspect workspace prerequisites before launching static audit:
 
 ```markdown
 ### 🔎 TorusGuard Static Audit Results
-- **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
+- **Confidence:** [Score]/100 (Evidence-Chain Calibrated)
 - **Living Report:** `security_report.md`
 ```
 
@@ -78,4 +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/payload/workflows/container.md

@@ -1,29 +1,29 @@
-# /torusguard-container — Container & Docker Security Audit
-
-$ARGUMENTS
-
----
-
-## Objective
-Detect and remediate container privilege escalation, root execution, docker socket exposure, and build-time secret injection across `Dockerfile`, `Containerfile`, and `docker-compose.yml` configurations.
-
----
-
-## Tri-Mode Parity
-- **Mode A (Terminal CLI):** `torusguard container [target]`
-- **Mode B (AI Chat Slash Command):** `/torusguard container [target]`
-- **Mode C (Native MCP Tool):** `torusguard_container`
-
----
-
-## Execution Steps
-
-1. **Discover Container Configurations:** Discover all Dockerfiles and compose YAML files in the workspace.
-2. **Execute Static Heuristic Audit:** Run `ScanContainerFiles` or `torusguard_container` against discovered files.
-3. **Assert Security Invariants:**
-   - Enforce non-root user declarations (`TG-CONT-001`).
-   - Block Docker daemon socket mounts (`/var/run/docker.sock`) (`TG-CONT-002`).
-   - Disallow `privileged: true` and unconfined seccomp profiles (`TG-CONT-003`).
-   - Ban build-time secrets in `ARG` or `ENV` directives (`TG-CONT-004`).
-4. **Formulate Surgical Patches:** Propose minimal diffs adding non-root users (`USER appuser`) and secret mounts within Ponytail bounds (≤35 additions, ≤25 deletions).
-5. **Synchronize Report:** Log all container findings in `security_report.md`.
+# /torusguard-container — Container & Docker Security Audit
+
+$ARGUMENTS
+
+---
+
+## Objective
+Detect and remediate container privilege escalation, root execution, docker socket exposure, and build-time secret injection across `Dockerfile`, `Containerfile`, and `docker-compose.yml` configurations.
+
+---
+
+## Tri-Mode Parity
+- **Mode A (Terminal CLI):** `torusguard container [target]`
+- **Mode B (AI Chat Slash Command):** `/torusguard container [target]`
+- **Mode C (Native MCP Tool):** `torusguard_container`
+
+---
+
+## Execution Steps
+
+1. **Discover Container Configurations:** Discover all Dockerfiles and compose YAML files in the workspace.
+2. **Execute Static Heuristic Audit:** Run `ScanContainerFiles` or `torusguard_container` against discovered files.
+3. **Assert Security Invariants:**
+   - Enforce non-root user declarations (`TG-CONT-001`).
+   - Block Docker daemon socket mounts (`/var/run/docker.sock`) (`TG-CONT-002`).
+   - Disallow `privileged: true` and unconfined seccomp profiles (`TG-CONT-003`).
+   - Ban build-time secrets in `ARG` or `ENV` directives (`TG-CONT-004`).
+4. **Formulate Surgical Patches:** Propose minimal diffs adding non-root users (`USER appuser`) and secret mounts within Ponytail bounds (≤35 additions, ≤25 deletions).
+5. **Synchronize Report:** Log all container findings in `security_report.md`.

package/skills/torusguard/payload/workflows/git-mine.md

@@ -1,25 +1,25 @@
-# /torusguard-git-mine — Git Repository & History Secret Mining
-
-$ARGUMENTS
-
----
-
-## Objective
-Inspect local Git metadata, `.git/config` remotes, and historical commit logs to uncover leaked API tokens, private certificates, and sensitive credentials committed in previous revisions.
-
----
-
-## Tri-Mode Parity
-- **Mode A (Terminal CLI):** `torusguard git-mine [target]`
-- **Mode B (AI Chat Slash Command):** `/torusguard git-mine [target]`
-- **Mode C (Native MCP Tool):** `torusguard_git_mine`
-
----
-
-## Execution Steps
-
-1. **Verify Git Environment:** Verify target directory is a valid Git repository containing `.git/`.
-2. **Audit Remote Configurations:** Check `.git/config` for embedded basic auth credentials in remote origin URLs (`TG-GIT-002`).
-3. **Mine Commit Log Diffs:** Inspect the last 50 commits (`git log -p`) for added tokens, keys, and credentials (`TG-GIT-001`).
-4. **Audit Tracked Sensitive Files:** Ensure `.env`, private keys, and certificates are not tracked in git tree history (`TG-GIT-003`).
-5. **Report & Revoke:** Provide immediate provider revocation steps and commit log cleanup guidance in `security_report.md`.
+# /torusguard-git-mine — Git Repository & History Secret Mining
+
+$ARGUMENTS
+
+---
+
+## Objective
+Inspect local Git metadata, `.git/config` remotes, and historical commit logs to uncover leaked API tokens, private certificates, and sensitive credentials committed in previous revisions.
+
+---
+
+## Tri-Mode Parity
+- **Mode A (Terminal CLI):** `torusguard git-mine [target]`
+- **Mode B (AI Chat Slash Command):** `/torusguard git-mine [target]`
+- **Mode C (Native MCP Tool):** `torusguard_git_mine`
+
+---
+
+## Execution Steps
+
+1. **Verify Git Environment:** Verify target directory is a valid Git repository containing `.git/`.
+2. **Audit Remote Configurations:** Check `.git/config` for embedded basic auth credentials in remote origin URLs (`TG-GIT-002`).
+3. **Mine Commit Log Diffs:** Inspect the last 50 commits (`git log -p`) for added tokens, keys, and credentials (`TG-GIT-001`).
+4. **Audit Tracked Sensitive Files:** Ensure `.env`, private keys, and certificates are not tracked in git tree history (`TG-GIT-003`).
+5. **Report & Revoke:** Provide immediate provider revocation steps and commit log cleanup guidance in `security_report.md`.

package/skills/torusguard/payload/workflows/ocr-scan.md

@@ -1,25 +1,25 @@
-# /torusguard-ocr-scan — Multi-Modal OCR Secret Scanning
-
-$ARGUMENTS
-
----
-
-## Objective
-Detect optical leaks of private API keys, AWS credentials, database URIs, GitHub personal access tokens, and private certificates embedded inside visual artifacts (`.png`, `.jpg`, `.jpeg`, `.webp`, `.bmp`, `.tiff`) up to 10MB using Tesseract OCR and deterministic regex signatures.
-
----
-
-## Tri-Mode Parity
-- **Mode A (Terminal CLI):** `torusguard ocr-scan <target>`
-- **Mode B (AI Chat Slash Command):** `/torusguard ocr-scan <target>`
-- **Mode C (Native MCP Tool):** `torusguard_ocr_scan`
-
----
-
-## Execution Steps
-
-1. **Verify Target:** Check target image or directory exists on disk.
-2. **Inspect File Size:** Reject images exceeding 10MB (`DefaultMaxImageSize`) to preserve DoS resilience.
-3. **Execute OCR Extraction:** Run Tesseract optical text analysis via Go scanner or MCP tool.
-4. **Pattern Match Secrets:** Evaluate optical text against `TG-SEC-001` through `TG-SEC-007` signatures.
-5. **Redact & Report:** Emit sanitized finding cards and synchronize into `security_report.md`.
+# /torusguard-ocr-scan — Multi-Modal OCR Secret Scanning
+
+$ARGUMENTS
+
+---
+
+## Objective
+Detect optical leaks of private API keys, AWS credentials, database URIs, GitHub personal access tokens, and private certificates embedded inside visual artifacts (`.png`, `.jpg`, `.jpeg`, `.webp`, `.bmp`, `.tiff`) up to 10MB using Tesseract OCR and deterministic regex signatures.
+
+---
+
+## Tri-Mode Parity
+- **Mode A (Terminal CLI):** `torusguard ocr-scan <target>`
+- **Mode B (AI Chat Slash Command):** `/torusguard ocr-scan <target>`
+- **Mode C (Native MCP Tool):** `torusguard_ocr_scan`
+
+---
+
+## Execution Steps
+
+1. **Verify Target:** Check target image or directory exists on disk.
+2. **Inspect File Size:** Reject images exceeding 10MB (`DefaultMaxImageSize`) to preserve DoS resilience.
+3. **Execute OCR Extraction:** Run Tesseract optical text analysis via Go scanner or MCP tool.
+4. **Pattern Match Secrets:** Evaluate optical text against `TG-SEC-001` through `TG-SEC-007` signatures.
+5. **Redact & Report:** Emit sanitized finding cards and synchronize into `security_report.md`.

package/skills/torusguard/payload/workflows/redos.md

@@ -1,27 +1,27 @@
-# /torusguard-redos — Regular Expression Complexity Scanner
-
-$ARGUMENTS
-
----
-
-## Objective
-Detect and remediate catastrophic exponential ($O(2^n)$) and polynomial backtracking regular expressions in web applications to prevent availability and thread starvation DoS attacks.
-
----
-
-## Tri-Mode Parity
-- **Mode A (Terminal CLI):** `torusguard redos [target]`
-- **Mode B (AI Chat Slash Command):** `/torusguard redos [target]`
-- **Mode C (Native MCP Tool):** `torusguard_redos`
-
----
-
-## Execution Steps
-
-1. **Locate Regex Definitions:** Walk source tree to find regex literals and constructor calls (`RegExp`, `re.compile`, `regexp.MustCompile`).
-2. **Analyze Backtracking Complexity:**
-   - Detect nested quantifiers (`(x+)+`, `(x*)*`) causing $O(2^n)$ exponential blowup (`TG-REDOS-001`).
-   - Detect overlapping alternations with outer repetition causing polynomial blowup (`TG-REDOS-002`).
-3. **Audit Input Length Guardrails:** Check if untrusted inputs evaluated against regexes are bounded in character length.
-4. **Formulate Safe Remediation:** Refactor vulnerable regexes into linear deterministic expressions, anchor tokens, or parser logic.
-5. **Synchronize Report:** Log all ReDoS findings in `security_report.md`.
+# /torusguard-redos — Regular Expression Complexity Scanner
+
+$ARGUMENTS
+
+---
+
+## Objective
+Detect and remediate catastrophic exponential ($O(2^n)$) and polynomial backtracking regular expressions in web applications to prevent availability and thread starvation DoS attacks.
+
+---
+
+## Tri-Mode Parity
+- **Mode A (Terminal CLI):** `torusguard redos [target]`
+- **Mode B (AI Chat Slash Command):** `/torusguard redos [target]`
+- **Mode C (Native MCP Tool):** `torusguard_redos`
+
+---
+
+## Execution Steps
+
+1. **Locate Regex Definitions:** Walk source tree to find regex literals and constructor calls (`RegExp`, `re.compile`, `regexp.MustCompile`).
+2. **Analyze Backtracking Complexity:**
+   - Detect nested quantifiers (`(x+)+`, `(x*)*`) causing $O(2^n)$ exponential blowup (`TG-REDOS-001`).
+   - Detect overlapping alternations with outer repetition causing polynomial blowup (`TG-REDOS-002`).
+3. **Audit Input Length Guardrails:** Check if untrusted inputs evaluated against regexes are bounded in character length.
+4. **Formulate Safe Remediation:** Refactor vulnerable regexes into linear deterministic expressions, anchor tokens, or parser logic.
+5. **Synchronize Report:** Log all ReDoS findings in `security_report.md`.

package/skills/torusguard/payload/workflows/review.md

@@ -0,0 +1,11 @@
+# TorusGuard Review Workflow (`/torusguard review`)
+
+## Purpose
+Executes differential incremental Git diff security review against a reference branch or commit (default: `HEAD~1`).
+Inspects changed lines for new security regressions, evaluates net security score deltas, and outputs pull request review comments.
+
+## Steps
+1. Determine reference branch or commit (e.g. `HEAD~1` or `origin/main`).
+2. Run differential scan via `torusguard review [--diff <ref>]` or native MCP tool `torusguard_review`.
+3. Check PR Gate Decision (`PASSED` or `BLOCKED`).
+4. If blocked, review newly introduced violations and formulate surgical Ponytail fixes via `/torusguard harden`.

package/skills/torusguard/payload/workflows/threatmodel.md

@@ -0,0 +1,9 @@
+# TorusGuard Threat Model Workflow (`/torusguard threatmodel`)
+
+## Purpose
+Synthesizes architectural STRIDE threat models, interactive Mermaid Data Flow Diagrams (DFDs), and trust boundary registers directly into `SECURITY_THREAT_MODEL.md`.
+
+## Steps
+1. Execute `torusguard threatmodel` or native MCP tool `torusguard_threatmodel`.
+2. Inspect `SECURITY_THREAT_MODEL.md` to verify identified entrypoints, data stores, and trust boundaries.
+3. Align STRIDE risk controls against corresponding TorusGuard security invariants (`TG-AUTH`, `TG-DB`, `TG-RATE`, `TG-SSRF`).