torusguard 1.3.3 → 1.3.5

package/skills/torusguard/payload/skills/torusguard-web-validate/SKILL.md

@@ -1,10 +1,11 @@
 ---
 name: torusguard-web-validate
 description: Execute authorized HTTP probing against local/staging web applications — session capture, transparent audit headers, and secret redaction.
-version: 0.9.2
+version: 1.3.5
 workflow: .torusguard/workflows/web-validate.md
-tools: Read, Grep, Glob, Bash, Write
+tools: Read, Grep, Glob, Bash, Write, run_command
 scripts-binding:
+  - .torusguard/scripts/report_sync.py
   - .torusguard/scripts/safety_gate.py
 ---
 
@@ -34,11 +35,12 @@ All requests pass through `.torusguard/scripts/safety_gate.py`:
 ## Execution Steps
 
 1. **Gate Check:** Validate scope in `scope.json` and query `safety_gate.py`.
-2. **Inject Audit Header:** Attach `X-TorusGuard-AuthID: <auth_id>` to all outbound requests.
-3. **Dispatch Bounded Probe:** Send single non-destructive HTTP request.
-4. **Validate Security Headers:** Check for CSP, HSTS, X-Content-Type-Options, and CORS configurations.
-5. **Audit Cookie Flags:** Verify `HttpOnly`, `Secure`, and `SameSite` on session cookies.
-6. **Capture Replay Trace:** Save sanitized trace in `.torusguard/runs/<run_id>/requests.json`.
+2. **Handle Automation Failure:** If automated web validation scripts fail, YOU must manually construct the HTTP requests (using `curl` or `fetch`) and inspect the responses.
+3. **Inject Audit Header:** Attach `X-TorusGuard-AuthID: <auth_id>` to all outbound requests.
+4. **Dispatch Bounded Probe:** Send single non-destructive HTTP request.
+5. **Validate Security Headers:** Check for CSP, HSTS, X-Content-Type-Options, and CORS configurations.
+6. **Audit Cookie Flags:** Verify `HttpOnly`, `Secure`, and `SameSite` on session cookies.
+7. **Capture Replay Trace:** Save sanitized trace in `.torusguard/runs/<run_id>/requests.json`.
 
 ---
 
@@ -47,6 +49,10 @@ All Bearer tokens, cookies, passwords, and API keys are redacted prior to disk s
 
 ---
 
+## Living Report Ground Truth
+- Always inspect `security_report.md` at workspace root before acting.
+- Update finding statuses after completion to eliminate hallucination.
+
 ## Safety Constraints
 - Max 50 requests per validation session.
 - Never test unauthorized domains or third-party APIs.
@@ -56,7 +62,7 @@ All Bearer tokens, cookies, passwords, and API keys are redacted prior to disk s
 
 ## Output Format
 ```markdown
-🌐 [TorusGuard] Web Validation Completed
+🌐 [TorusGuard] Web Validation Completed (AI Assisted)
 - Target: <Host URL> | Endpoints Tested: <Count>
 - Header Posture: <CSP/HSTS Status> | Cookies: <Flags Status>
 - Replay Trace: `.torusguard/runs/<run_id>/requests.json`

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

@@ -1,12 +1,14 @@
 ---
 description: Governed patch application with pre-apply rollback snapshots and Human Gate authorization.
 tools: Read, Grep, Glob, Bash, Edit, Write
-version: 0.9.2
+version: 1.3.5
 agent: remediator
 lifecycle-phase: Phase 5 (Patch Application)
 required-skills:
   - torusguard-apply
 scripts-binding:
+  - .torusguard/scripts/apply_runner.py
+  - .torusguard/scripts/report_sync.py
   - .torusguard/scripts/run_manager.py
   - .torusguard/scripts/diff_guard.py
 ---
@@ -26,7 +28,7 @@ Governed patch application with pre-apply rollback snapshots and Human Gate auth
 
 Inspect bundle validity and file safety before modifying disk code:
 1. **Bundle Verification:** Assert `patch.diff` exists in active run's remediation folder.
-2. **Pre-Apply Snapshot:** Save a byte-for-byte backup copy (`.bak` or `pre_apply/`) prior to editing.
+2. **Pre-Apply Snapshot:** Save a byte-for-byte backup copy (`.bak` or `.torusguard/snapshots/<run_id>/`) prior to editing.
 3. **Uncommitted Changes:** Check git status; ensure target file has clean baseline.
 4. **Ponytail Check:** Re-verify that patch additions $\le 35$ and deletions $\le 25$.
 5. **Human Gate:** Confirm operator approval before committing edits to disk.
@@ -47,11 +49,16 @@ Inspect bundle validity and file safety before modifying disk code:
 
 ---
 
+## Living Report Invariant
+- Applied patches create pre-apply `.bak` snapshots in `.torusguard/snapshots/<run_id>/`.
+- Finding statuses transition in `security_report.md` to `APPLIED 🔵`.
+- Instant rollback available via `npx torusguard rollback`.
+
 ## Execution Steps
 
 1. **Load Remediation Bundle:** Read `patch.diff` and metadata from active run directory.
 2. **Audit Patch Invariants:** Run `python .torusguard/scripts/diff_guard.py <patch.diff>`.
-3. **Create Rollback Backup:** Copy target file to `.torusguard/runs/<run_id>/pre_apply/<filename>.bak`.
+3. **Create Rollback Backup:** Copy target file to `.torusguard/runs/<run_id>/.torusguard/snapshots/<run_id>/<filename>.bak`.
 4. **Apply Minimal Diff:** Execute precise surgical edit using `replace_file_content` or `patch`.
 5. **Assert Syntax & Integrity:** Check that modified file compiles cleanly without syntax errors.
 6. **Log Application Ledger:** Record timestamp, original SHA-256, and patched SHA-256 in `apply-log.json`.
@@ -81,7 +88,7 @@ Inspect bundle validity and file safety before modifying disk code:
 ### 🚀 TorusGuard Patch Application
 - **Finding Addressed:** `TG-XXX-HASH`
 - **File Patched:** `src/path/to/file.py`
-- **Rollback Backup:** `.torusguard/runs/<run_id>/pre_apply/<file>.bak`
+- **Rollback Backup:** `.torusguard/runs/<run_id>/.torusguard/snapshots/<run_id>/<file>.bak`
 - **Churn Applied:** +[Additions] / -[Deletions] lines
 - **Syntax Check:** PASSED (clean compile)
 - **Status:** APPLIED — run `/torusguard recheck` to verify fix

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

@@ -1,12 +1,14 @@
 ---
 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: 0.9.2
+version: 1.3.5
 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
 ---
@@ -45,6 +47,10 @@ Inspect workspace prerequisites before launching static audit:
 
 ---
 
+## Living Report Invariant
+- 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`.

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

@@ -1,12 +1,13 @@
 ---
 description: Legal scope definition, target ownership proof verification, and safety boundary enforcement for runtime validation.
 tools: Read, Grep, Glob, Bash, Write
-version: 0.9.2
+version: 1.3.5
 agent: reviewer
 lifecycle-phase: Phase 1 (Authorization Gate)
 required-skills:
   - torusguard-authorize
 scripts-binding:
+  - .torusguard/scripts/report_sync.py
   - .torusguard/scripts/safety_gate.py
 ---
 

package/skills/torusguard/payload/workflows/exploit-check.md

@@ -1,12 +1,13 @@
 ---
 description: Safe, bounded, single-step exploitability confirmation across 5 standard verification statuses.
 tools: Read, Grep, Glob, Bash, Write
-version: 0.9.2
+version: 1.3.5
 agent: validator
 lifecycle-phase: Phase 3c (Exploitability Check)
 required-skills:
   - torusguard-exploit-check
 scripts-binding:
+  - .torusguard/scripts/report_sync.py
   - .torusguard/scripts/safety_gate.py
   - .torusguard/scripts/finding_scorer.py
 ---

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

@@ -1,12 +1,14 @@
 ---
 description: Governed remediation formulation under strict Ponytail Protocol bounds (<= 35 additions, <= 25 deletions) and bundle packaging.
 tools: Read, Grep, Glob, Bash, Write
-version: 0.9.2
+version: 1.3.5
 agent: remediator
 lifecycle-phase: Phase 4 (Remediation Formulation)
 required-skills:
   - torusguard-harden
 scripts-binding:
+  - .torusguard/scripts/harden_runner.py
+  - .torusguard/scripts/report_sync.py
   - .torusguard/scripts/run_manager.py
   - .torusguard/scripts/diff_guard.py
 ---
@@ -46,6 +48,10 @@ Inspect finding targets and patch constraints before generating diffs:
 
 ---
 
+## Living Report Invariant
+- Candidate patches transition findings in `security_report.md` into `CANDIDATE 🟡` status.
+- Every patch strictly conforms to Ponytail bounds (<= 35 additions, <= 25 deletions).
+
 ## Execution Steps
 
 1. **Select Target Finding:** Choose verified flaw from active run directory.

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

@@ -1,12 +1,13 @@
 ---
 description: Baseline project discovery, workspace scaffolding, stack detection, and framework-tailored security rule activation.
 tools: Read, Grep, Glob, Bash, Write
-version: 0.9.2
+version: 1.3.5
 agent: profiler
 lifecycle-phase: Phase 0 (Baseline Setup)
 required-skills:
   - torusguard-init
 scripts-binding:
+  - .torusguard/scripts/report_sync.py
   - .torusguard/scripts/stack_detect.py
 ---
 
@@ -44,6 +45,10 @@ Inspect workspace state before running initialization:
 
 ---
 
+## Living Report Invariant
+- Provisions baseline `security_report.md` single-source-of-truth ledger.
+- Activates tailored rules across 18 families based on detected stack.
+
 ## Execution Steps
 
 1. **Scaffold Directory Topology:** Verify `.torusguard/` structure (`config/`, `rules/active/`, `runs/`, `scripts/`, `workflows/`, `templates/`, `schemas/`).

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

@@ -1,12 +1,13 @@
 ---
 description: TorusGuard Adaptive Security Memory Engine — view memory stats, export/import intelligence, apply TTL decay, and manage false positives.
 tools: Read, Grep, Glob, Bash
-version: 1.0.0
+version: 1.3.5
 agent: reviewer
 lifecycle-phase: Intelligence / Memory
 required-skills:
   - torusguard
 scripts-binding:
+  - .torusguard/scripts/report_sync.py
   - .torusguard/scripts/memory_engine.py
 ---
 

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

@@ -1,12 +1,14 @@
 ---
 description: Targeted differential AST re-scan, fix closure verification, and regression state machine transitions.
 tools: Read, Grep, Glob, Bash, Write
-version: 0.9.2
+version: 1.3.5
 agent: reviewer
 lifecycle-phase: Phase 6 (Targeted Recheck & Verification)
 required-skills:
   - torusguard-recheck
 scripts-binding:
+  - .torusguard/scripts/recheck_runner.py
+  - .torusguard/scripts/report_sync.py
   - .torusguard/scripts/run_manager.py
   - .torusguard/scripts/finding_scorer.py
 ---
@@ -46,6 +48,10 @@ Inspect applied patch history and target file scope before re-checking:
 
 ---
 
+## Living Report Invariant
+- Verified fix closures transition findings in `security_report.md` to `RESOLVED 🟢`.
+- Regressions transition findings to `REGRESSED ❌`.
+
 ## Execution Steps
 
 1. **Identify Modified Files:** Extract list of altered files from latest apply step.

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

@@ -1,12 +1,14 @@
 ---
 description: Executive posture reporting, cluster analysis, signed compliance audit, and OASIS SARIF v2.1.0 structured export.
 tools: Read, Grep, Glob, Bash, Write
-version: 0.9.2
+version: 1.3.5
 agent: reviewer
 lifecycle-phase: Phase 7 (Reporting & SARIF Export)
 required-skills:
   - torusguard-report
 scripts-binding:
+  - .torusguard/scripts/html_reporter.py
+  - .torusguard/scripts/report_sync.py
   - .torusguard/scripts/sarif_exporter.py
   - .torusguard/scripts/run_manager.py
 ---
@@ -45,6 +47,10 @@ Inspect run records and reporting parameters before generating release artifacts
 
 ---
 
+## Living Report Invariant
+- Executive posture reports incorporate health scoring and metrics from `security_report.md`.
+- Emits visual dark-mode HTML (`--html`) and OASIS SARIF v2.1.0 (`--sarif`).
+
 ## Execution Steps
 
 1. **Aggregate Lifecycle Data:** Load all finding cards, verification traces, and recheck logs.

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

@@ -1,12 +1,13 @@
 ---
 description: Read-only diagnostic overview of workspace security posture, active configuration, rules catalog, and run history.
 tools: Read, Grep, Glob, Bash
-version: 0.9.2
+version: 1.3.5
 agent: reviewer
 lifecycle-phase: System / Diagnostics
 required-skills:
   - torusguard-status
 scripts-binding:
+  - .torusguard/scripts/report_sync.py
   - .torusguard/scripts/run_manager.py
 ---
 
@@ -44,6 +45,10 @@ Inspect workspace state and configuration records prior to status display:
 
 ---
 
+## Living Report Invariant
+- Reads current health score and status breakdown directly from `security_report.md`.
+- Displays 18 rule families and 74 rules in standardized 75-column terminal cards.
+
 ## Execution Steps
 
 1. **Read Configuration:** Load `.torusguard/config/torusguard.json`; extract stack, version, and enabled rule count.

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

@@ -1,12 +1,13 @@
 ---
 description: Evidence sufficiency verification, live disk line match audit, and finding score refinement.
 tools: Read, Grep, Glob, Bash, Write
-version: 0.9.2
+version: 1.3.5
 agent: validator
 lifecycle-phase: Phase 3a (Evidence Verification)
 required-skills:
   - torusguard-verify
 scripts-binding:
+  - .torusguard/scripts/report_sync.py
   - .torusguard/scripts/finding_scorer.py
 ---
 
@@ -44,6 +45,10 @@ Inspect finding records and evidence integrity before verification:
 
 ---
 
+## Living Report Invariant
+- Confirmed evidence sufficiency transitions findings in `security_report.md` to `VERIFIED 🟠`.
+- False positives transition to `FALSE POSITIVE ⚪`.
+
 ## Execution Steps
 
 1. **Locate Target Findings:** Load active findings from latest run directory.

package/skills/torusguard/payload/workflows/web-validate.md

@@ -1,12 +1,13 @@
 ---
 description: Authorized HTTP probing, token redaction, transparent audit header injection, and replay trace capture.
 tools: Read, Grep, Glob, Bash, Write
-version: 0.9.2
+version: 1.3.5
 agent: validator
 lifecycle-phase: Phase 3b (Runtime Web Validation)
 required-skills:
   - torusguard-web-validate
 scripts-binding:
+  - .torusguard/scripts/report_sync.py
   - .torusguard/scripts/safety_gate.py
 ---
 

package/skills/torusguard-apply/SKILL.md

@@ -1,48 +1,84 @@
 ---
 name: torusguard-apply
-description: Apply governed remediation patches to disk with pre-apply rollback snapshots and Human Gate validation.
-version: 0.9.2
+description: Apply governed remediation patches to disk with pre-apply rollback snapshots and Human Gate validation via CLI or AI Agent.
+version: 1.3.5
 workflow: .torusguard/workflows/apply.md
-tools: Read, Grep, Glob, Bash, Edit, Write
+tools: Read, Grep, Glob, Write, replace_file_content, run_command
 scripts-binding:
-  - .torusguard/scripts/run_manager.py
-  - .torusguard/scripts/diff_guard.py
+  - .torusguard/scripts/report_sync.py
+  - .torusguard/scripts/apply_runner.py
+  - .torusguard/scripts/term_ui.py
+  - .torusguard/scripts/memory_engine.py
 ---
 
 # TorusGuard Apply — Governed Patch Application & Rollback Snapshot
 
 ## Objective
-Safely apply approved remediation bundles to disk source files, capturing a byte-for-byte pre-apply backup snapshot and asserting syntax validity post-patch.
+Safely apply approved remediation bundles to disk source files, creating an automated byte-for-byte pre-apply backup snapshot (`.bak`) in `.torusguard/snapshots/<run_id>/`, asserting syntax validity, and distilling passing patterns into Golden Fix Recipes.
 
 ---
 
-## Execution Steps
+## Two Execution Modes
 
-1. **Load Bundle:** Read `patch.diff` and metadata from `.torusguard/runs/<run_id>/remediation/<finding_id>/`.
-2. **Handle CLI Failures:** If the CLI `npx torusguard apply` throws an error or fails to apply the patch, YOU MUST step in and apply it manually.
-3. **Pre-Apply Snapshot:** Save exact backup to `.torusguard/runs/<run_id>/pre_apply/<filename>.bak`.
-4. **Obtain Human Gate:** Confirm user approval for proposed diff.
-5. **Apply Patch Manually:** Execute the surgical modification directly on the target file using `replace_file_content`. Ensure the lines match the `patch.diff` formulated during harden phase.
-6. **Assert Integrity:** Verify modified file compiles cleanly without syntax errors (e.g. running `npm run build` or `python -m py_compile`).
-7. **Record Ledger:** Log timestamp, file path, and post-patch SHA-256 in `apply-log.json`.
+### Mode A: Automated CLI Execution (Interactive or Non-Interactive)
+Run the governed patch applier from your terminal:
+```bash
+# Interactive Human Gate review (prompts for each patch: [y]es / [n]o / [a]ll / [q]uit)
+npx torusguard apply
+
+# Non-interactive automated application
+npx torusguard apply --yes
+
+# Apply patches from a specific run ID
+npx torusguard apply --run run-20260910-121618-audit --yes
+
+# Emergency rollback from pre-apply snapshot
+npx torusguard rollback
+# or
+npx torusguard apply --rollback
+```
+**Under the Hood:** Executes `python .torusguard/scripts/apply_runner.py`.
+1. Reads candidate bundles from `.torusguard/runs/<run_id>/bundles/`.
+2. Creates `.bak` snapshot in `.torusguard/snapshots/<run_id>/<rel_path>.bak`.
+3. Performs line-aware replacement without corrupting concurrent modifications.
+4. Distills the applied fix into a **Golden Fix Recipe** registered in `.torusguard/memory/patterns.json`.
+5. Emits `diff_summary.md` and `apply_plan.md` in the run directory.
+6. Displays pixel-perfect 75-column terminal cards.
+
+### Mode B: In-Session AI Chat Agent Application
+When applying patches directly within an AI chat session:
+1. **Human Gate Confirmation:** Present the unified diff to the operator and request explicit approval before editing code.
+2. **Pre-Apply Snapshot:** Ensure target file backup is created at `.torusguard/snapshots/<run_id>/<rel_path>.bak`.
+3. **Execute Surgical Edit:** Use `replace_file_content` to apply only the approved lines. Strictly adhere to the formulated Ponytail diff.
+4. **Syntax & Integrity Check:** Verify that the patched file is syntactically valid (e.g. `node --check <file>` or `python -m py_compile <file>`). If syntax fails, immediately restore from `.bak`.
+5. **Memory Registration:** Update persistent memory events in `.torusguard/memory/events.json` with event `fix_applied`.
+6. **Recommend Recheck:** Advise running `/torusguard recheck` or `npx torusguard recheck` to verify closure.
 
 ---
 
-## Governance & Safety Rules
-- Hard limit: additions $\le 35$, deletions $\le 25$ lines.
-- Always create a rollback snapshot before modifying disk code.
-- If compile or syntax error occurs, instantly restore from `.bak`.
-- Require explicit operator confirmation before writing changes.
-- **Do not give up** if the CLI automation fails; the AI Agent must manually complete the patching process.
+## Rollback Guarantee
+If an applied patch introduces unexpected runtime behavior or breaks tests:
+```bash
+# Instant one-command rollback of the latest applied run:
+npx torusguard rollback
+
+# Or rollback a specific run:
+npx torusguard rollback --run run-20260910-121618-audit
+```
+All affected files are restored byte-for-byte from `.torusguard/snapshots/<run_id>/`.
 
 ---
 
-## Output Format
+## Output Card Format
 ```markdown
-🚀 [TorusGuard] Patch Applied Successfully (AI Assisted)
-- Finding Target: <Finding ID> | File: <Path>
-- Churn: +<Additions> / -<Deletions>
-- Rollback Backup: `.torusguard/runs/<run_id>/pre_apply/<file>.bak`
-- Compile Check: PASSED
-Next: Run `/torusguard recheck` to verify vulnerability is closed.
+### 🚀 TorusGuard Patch Applied Successfully
+- **Target File:** `server/index.js:9`
+- **Rule ID:** `TG-SEC-001` (Hardcoded JWT secret)
+- **Ponytail Churn:** +1 / -1
+- **Rollback Snapshot:** `.torusguard/snapshots/<run_id>/server/index.js.bak`
+- **Golden Recipe:** Distilled into persistent memory
+- **Next Step:** Run `npx torusguard recheck` or `/torusguard recheck` to verify closure
 ```
+
+## Living Report Ground Truth
+- Read `security_report.md` in the workspace root before taking any action. Update the relevant finding card after completing remediation.

package/skills/torusguard-audit/SKILL.md

@@ -1,75 +1,87 @@
 ---
 name: torusguard-audit
-description: Static AST scanning, memory pattern boosting, and 0–100 confidence scoring.
-version: 1.0.0
+description: Static AST security scanning, line-shift invariant fingerprinting, root-cause clustering, and 0-100 confidence scoring via CLI or AI Agent.
+version: 1.3.5
 workflow: .torusguard/workflows/audit.md
-tools: Read, Grep, Glob, Write
+tools: Read, Grep, Glob, Write, run_command
 scripts-binding:
+  - .torusguard/scripts/report_sync.py
+  - .torusguard/scripts/audit_runner.py
+  - .torusguard/scripts/term_ui.py
   - .torusguard/scripts/finding_scorer.py
-  - .torusguard/scripts/run_manager.py
 ---
 
-# TorusGuard Audit — Static Code Analysis & Memory Scanning
+# TorusGuard Audit — Static Code Security Analysis
 
 ## Objective
-Execute static AST analysis, query memory context, compute 0–100 confidence scores with memory boosts, and record findings to runs and memory.
+Execute static AST analysis across polyglot project files, evaluate code against 74 canonical security rules across 11 families, assign stable line-shift invariant fingerprints, cluster architectural root causes, and score findings with auditable 0–100 confidence ratings.
 
 ---
 
-## Execution Steps
+## Two Execution Modes
 
-1. **Run Setup:** Initialize `.torusguard/runs/run-YYYYMMDD-HHMMSS-audit/` via `run_manager.py`.
-2. **Context & Rules:** Load `.torusguard/memory/context.json` for patterns and false positives. Load rules from `.torusguard/rules/active/`.
-3. **AST Scan Sinks:** Match files against active rules:
-   - `TG-SEC-*`: Secrets and API keys.
-   - `TG-INPUT-*`: SQL injection, unvalidated input.
-   - `TG-DB-*`: Unscoped tenant queries (`Model.get(id=id)`).
-   - `TG-AUTH-*`: Insecure cookies, open CORS.
-   - `TG-CLIENT-*`: Client bundle secret leaks.
-4. **Stable Fingerprinting:** Compute SHA-256 hash of surrounding AST context.
-5. **Root-Cause Clustering:** Group findings by causal sink (`cluster-tenant-isolation`).
-6. **Scoring:** Calculate 0–100 score via `finding_scorer.py` (applies memory boost/suppression).
-7. **Write Artifacts:** Save `findings.json` and sync findings to memory engine.
+### Mode A: Automated CLI Execution
+Run the static security audit from your terminal:
+```bash
+# Scan current repository
+npx torusguard audit
 
----
-
-## Confidence Scoring Rubric
-- **1. Evidence (35):** AST flow (35), regex sink (20), heuristic (10).
-- **2. Reproduction (25):** Automated test (25), simulated (15), hypothesis (0).
-- **3. Corroboration (15):** >=3 files (15), 2 files (10), single (5).
-- **4. Environment (15):** Explicit route (15), ambiguous (8), unknown (0).
-- **5. Review Status (10):** Human confirmed (10), agent verified (5), auto (0).
-- **Memory Modifier:** False positive (-30), regression (+15), recurring (+10).
-- **Bands:** 90–100 (`Confirmed`), 70–89 (`High`), 50–69 (`Medium`), <50 (`Review`).
+# Scan specific directory or example app
+npx torusguard audit ./examples/vulnerable-react-express
 
----
+# Include test fixtures and spec directories
+npx torusguard audit --include-tests
 
-## Finding Card Format
-```markdown
-### [TG-DB-004] Missing Tenant Scoping in /invoices/
-- **Severity**: High | **Confidence**: 90/100 (Confirmed) | **Memory Boost**: +10
-- **File**: `views.py:42` | **Cluster**: `cluster-tenant-isolation`
-- **Remediation**: Apply `.filter(tenant=request.user.tenant)`.
+# Output machine-readable JSON
+npx torusguard audit --json
 ```
+**Under the Hood:** Executes `python .torusguard/scripts/audit_runner.py`.
+- Auto-detects repository stack and skips build/cache directories (`node_modules`, `.git`, `.venv`, `dist`, `build`).
+- Evaluates files across 18 canonical security families:
+  - `TG-SEC-*`: Hardcoded credentials, private keys, JWT secrets, client env leaks.
+  - `TG-INPUT-*`: SQL injection, command injection, path traversal, unsafe HTML rendering.
+  - `TG-DB-*`: Missing tenant isolation, service role keys in client code.
+  - `TG-AUTH-*`: Plaintext passwords, missing cookie security flags (httpOnly, secure, sameSite).
+  - `TG-PLATFORM-*`: Permissive wildcard CORS with credentials, missing security headers.
+  - `TG-DIFF-*`: Disabled TLS verification (`verify=False`, `rejectUnauthorized: false`).
+- Generates run directory: `.torusguard/runs/run-YYYYMMDD-HHMMSS-audit/`.
+- Writes `findings.json`, `findings.md`, and `manifest.json`.
+- Displays standardized 75-column terminal cards.
+- Returns exit code `1` if findings are discovered, `0` if clean.
+
+### Mode B: In-Session AI Chat Agent Scan
+When auditing files directly in AI chat:
+1. **Discover Sinks:** Use `grep_search` and `view_file` to search for dangerous patterns across server and client code.
+2. **Cluster Root Causes:** Group findings by causal architecture (e.g. `cluster-tenant-isolation`, `cluster-credentials-exposure`, `cluster-injection`).
+3. **Audit Evidence Sufficiency:** Ensure that user-controlled input reaches the vulnerable sink without prior sanitization or schema validation.
+4. **Present Actionable Findings:** Display finding cards with severity, rule ID, file, line, and remediation recommendation.
+5. **Prompt Next Phase:** Guide the operator to `/torusguard harden` or `npx torusguard harden`.
 
 ---
 
-## Safety Constraints
-- Read-only analysis; zero code modifications during audit.
-- Exclude build directories (`node_modules/`, `.venv/`, `dist/`).
-- Redact secrets before saving to `findings.md`.
+## Canonical Rule Families
+| Family | Scope | Example Violations |
+| :--- | :--- | :--- |
+| **TG-SEC** | Secrets & Credentials | Hardcoded JWT secret, API key strings, token logging |
+| **TG-INPUT** | Injection & Input Validation | Raw SQL interpolation, DOM `innerHTML`, `path.join` traversal |
+| **TG-DB** | Database & Tenant Scoping | Unscoped `.objects.get(id=...)`, Prisma missing `tenantId` |
+| **TG-AUTH** | Authentication & Cookies | Insecure cookies (missing httpOnly/secure/sameSite) |
+| **TG-PLATFORM** | Server & Platform Config | Wildcard CORS (`origin: '*'`) with credentials |
+| **TG-DIFF** | Security Bypasses | Disabled TLS verification (`verify=False`) |
 
 ---
 
-## Output Format
+## Output Card Format
 ```markdown
-🛡️ [TorusGuard] Static Code Audit Completed
-- Run ID: run-YYYYMMDD-HHMMSS-audit | Files: <Count> | Findings: <Count>
-- Posture Score: <Score>/100 (100 = Perfect, 0 = Extremely Vulnerable)
-- Clusters: <Count> identified | Memory Patterns Applied: <Count>
-
-> [!IMPORTANT]
-> If the posture score is low (e.g. 10/100) and vulnerabilities exist, it is your responsibility as the AI Agent to proactively recommend and manually execute the patches if the automated CLI falls short.
-
-Next: Run `/torusguard harden` to formulate surgical fixes.
+### 🛡️ TorusGuard Static Security Audit Completed
+- **Run ID:** `run-20260910-121618-audit`
+- **Scope:** 7 files evaluated across 11 canonical families
+- **Status:** ✖ CRITICAL FINDINGS DETECTED
+- **Findings:** 2 Critical, 2 High, 2 Medium/Low (6 total)
+- **Clusters:** 3 architectural root causes identified
+- **Artifacts:** `.torusguard/runs/<run_id>/findings.json`
+- **Next Action:** Run `npx torusguard harden` or `/torusguard harden`
 ```
+
+## Living Report Ground Truth
+- Read `security_report.md` in the workspace root before taking any action. Update the relevant finding card after completing remediation.

package/skills/torusguard-authorize/SKILL.md

@@ -1,10 +1,11 @@
 ---
 name: torusguard-authorize
 description: Register and validate runtime target authorization boundaries — scope boundaries, ownership proofs, TTL expiration, and Safety Gate enforcement.
-version: 0.9.2
+version: 1.3.5
 workflow: .torusguard/workflows/authorize.md
-tools: Read, Grep, Glob, Write
+tools: Read, Grep, Glob, Write, run_command
 scripts-binding:
+  - .torusguard/scripts/report_sync.py
   - .torusguard/scripts/safety_gate.py
 ---
 
@@ -30,6 +31,10 @@ Define and validate legal runtime authorization boundaries, verify target owners
 
 ---
 
+## Living Report Ground Truth
+- Always inspect `security_report.md` at workspace root before acting.
+- Update finding statuses after completion to eliminate hallucination.
+
 ## Safety Constraints
 - Never authorize wildcard hosts (`*`) or third-party domains.
 - State-changing destructive actions (`DELETE`, bulk drops) are disabled by default.