coverme-scanner 1.5.1 → 1.6.0

package/dist/prompts/coverme-command.md

@@ -1,1021 +1,250 @@
 # CoverMe - Ultimate AI Security Scanner
 
-The most comprehensive AI-powered code scanner. 22 specialized agents + validators + deep analysis.
+The most comprehensive AI-powered code scanner. 22 specialized agents + validators.
 
 $ARGUMENTS
 
-## CRITICAL INSTRUCTIONS - READ FIRST!
+## CRITICAL INSTRUCTIONS
 
-1. **DO NOT ASK ANY QUESTIONS** - Run the entire scan autonomously from start to finish
-2. **DO NOT STOP FOR CONFIRMATION** - Just keep going through all phases
-3. **DO NOT ASK ABOUT FILE CHANGES** - Automatically update/overwrite scan.json
-4. **DO NOT ASK TO OPEN REPORT** - Just open it automatically at the end
-5. **COMPLETE EVERYTHING IN ONE GO** - All phases without interruption
-6. **RUN AGENTS IN BACKGROUND** - Use `run_in_background: true` for all Task tool calls
-7. **RUN BASH IN BACKGROUND** - Use `run_in_background: true` for long Bash commands
-
-Execute ALL phases automatically. Do NOT stop until the HTML report is open.
+1. **DO NOT ASK ANY QUESTIONS** - Run autonomously
+2. **DO NOT STOP FOR CONFIRMATION** - Keep going
+3. **COMPLETE EVERYTHING** - All phases without interruption
+4. **AGENTS WRITE TO FILES** - Each agent writes results to `.coverme/agents/{ID}.json`
+5. **AGENTS RETURN ONLY "done" or "skipped"** - Never return findings in response
 
 ---
 
-## Phase 0: Project Discovery & Statistics
-
-### Step 1: Gather Project Statistics
+## Phase 0: Setup
 
 ```bash
-# Count files and lines of code
-find . -type f \( -name "*.ts" -o -name "*.js" -o -name "*.tsx" -o -name "*.jsx" -o -name "*.py" -o -name "*.go" -o -name "*.java" -o -name "*.rb" -o -name "*.php" -o -name "*.cs" -o -name "*.swift" -o -name "*.kt" \) -not -path "*/node_modules/*" -not -path "*/.git/*" -not -path "*/dist/*" -not -path "*/build/*" -not -path "*/__pycache__/*" | wc -l
-
-# Count lines of code (approximate)
-find . -type f \( -name "*.ts" -o -name "*.js" -o -name "*.tsx" -o -name "*.jsx" -o -name "*.py" -o -name "*.go" \) -not -path "*/node_modules/*" -not -path "*/.git/*" -not -path "*/dist/*" 2>/dev/null | head -100 | xargs wc -l 2>/dev/null | tail -1
-
-# Generate project tree (max 3 levels deep, exclude node_modules etc)
-find . -maxdepth 3 -type d -not -path "*/node_modules/*" -not -path "*/.git/*" -not -path "*/dist/*" -not -path "*/__pycache__/*" | head -30 | sort
+mkdir -p .coverme/agents
+rm -f .coverme/agents/*.json 2>/dev/null
 ```
 
-### Step 2: Read Project Info
-
+Get project stats:
 ```bash
-cat package.json 2>/dev/null | head -30
-cat README.md 2>/dev/null | head -100
-ls -la
-ls src/ 2>/dev/null || ls app/ 2>/dev/null || ls lib/ 2>/dev/null
-```
-
-### Step 3: Load Custom Agents (if exists)
-
-```bash
-cat .coverme/agents.json 2>/dev/null || echo "NO_CUSTOM_AGENTS"
-```
-
-### Step 4: Check for Runtime Verification (SSH)
-
-```bash
-cat .coverme/runtime.json 2>/dev/null || echo "NO_RUNTIME_CONFIG"
+FILES=$(find . -type f \( -name "*.ts" -o -name "*.js" -o -name "*.py" -o -name "*.go" \) -not -path "*/node_modules/*" -not -path "*/.git/*" -not -path "*/dist/*" 2>/dev/null | wc -l | tr -d ' ')
+LINES=$(find . -type f \( -name "*.ts" -o -name "*.js" -o -name "*.py" \) -not -path "*/node_modules/*" -not -path "*/dist/*" 2>/dev/null | head -50 | xargs wc -l 2>/dev/null | tail -1 | awk '{print $1}')
+echo "Files: $FILES, Lines: ~$LINES"
 ```
 
 ---
 
-## Phase 1: Discovery (22 parallel agents)
-
-Launch ALL agents IN PARALLEL using the Task tool with `run_in_background: true`.
-
-### Agent 1: Security Core Scanner (SEC)
-```
-Scan for OWASP Top 10 and common vulnerabilities:
-
-INJECTION:
-- SQL injection (string concatenation in queries, raw queries)
-- NoSQL injection (MongoDB $where, $regex with user input)
-- Command injection (exec, spawn, system with user input)
-- LDAP injection, XPath injection
-- Template injection (SSTI in Jinja2, EJS, Handlebars)
-- Header injection (CRLF in headers)
-- Log injection (unescaped user input in logs)
-
-XSS:
-- Reflected XSS (user input in response without encoding)
-- Stored XSS (database content rendered without escaping)
-- DOM XSS (innerHTML, document.write, eval with user data)
-- dangerouslySetInnerHTML in React without sanitization
-
-AUTHENTICATION:
-- Hardcoded credentials (check git ls-files first!)
-- Weak password policies (no complexity, short length)
-- Missing rate limiting on login/register
-- Session fixation (session ID not rotated after login)
-- JWT issues (none algorithm, weak secret, no expiry)
-- Missing MFA on sensitive operations
-
-AUTHORIZATION:
-- IDOR (direct object references without ownership check)
-- Missing authorization checks on endpoints
-- Privilege escalation paths
-- Horizontal access (user A accessing user B's data)
-- Vertical access (user accessing admin functions)
-
-CRYPTOGRAPHY:
-- MD5/SHA1 for passwords (use bcrypt/argon2)
-- Math.random() for security (use crypto.randomBytes)
-- Hardcoded encryption keys/IVs
-- ECB mode usage
-- Missing HTTPS enforcement
-
-DATABASE-SPECIFIC DANGEROUS FUNCTIONS:
-- DuckDB: read_text(), read_blob(), read_csv_auto(), read_parquet(), glob(), getenv(), httpfs
-- SQLite: load_extension(), readfile(), writefile()
-- PostgreSQL: pg_read_file(), pg_ls_dir(), COPY TO/FROM
-- MySQL: LOAD_FILE(), INTO OUTFILE, INTO DUMPFILE
-- MongoDB: $where with user input, mapReduce with user functions
-- Redis: EVAL/EVALSHA with user input, CONFIG, DEBUG commands
-
-Output JSON: [{id: "SEC-XXX", title, severity, category, file, line, code, description, impact, recommendation, cwe, confidence, fixOwner, fixType, dread: {damage, reproducibility, exploitability, affectedUsers, discoverability, score}}]
-```
-
-### Agent 2: Auth & Session Scanner (AUTH)
-```
-Deep dive into authentication and session management:
-
-FIRST: Detect which auth method(s) this project uses:
-| Auth Type | Detection Pattern |
-|-----------|-------------------|
-| OAuth/OIDC | oauth, oidc, authorization_code, client_id, redirect_uri |
-| JWT | jsonwebtoken, jwt, Bearer, accessToken, refreshToken |
-| Session-based | express-session, cookie-session, session.save, req.session |
-| API Keys | x-api-key, apiKey, api_key, API_SECRET |
-| Clerk | @clerk, useAuth, clerkMiddleware |
-| Auth0 | @auth0, auth0-js, auth0-react |
-| Firebase Auth | firebase/auth, signInWith, onAuthStateChanged |
-| Passport.js | passport, passport-local, passport-jwt |
-| Supabase Auth | @supabase/auth, supabase.auth |
-| NextAuth | next-auth, NextAuth, getServerSession |
-
-FOR EACH AUTH TYPE DETECTED, check specific vulnerabilities:
-
-OAuth/OIDC:
-- Open redirect in return_url/redirect_uri (CRITICAL!)
-- State parameter missing or predictable
-- PKCE not implemented for public clients
-- Token stored in localStorage (XSS vulnerable)
-
-JWT:
-- alg: none accepted (signature bypass)
-- Weak secret (< 256 bits)
-- No expiry (exp claim missing)
-- Token not invalidated on logout
-
-Session-based:
-- Session ID in URL (referer leak)
-- Session not invalidated on logout
-- Session fixation
-- Missing secure, httpOnly, sameSite on cookies
-
-TIMING ATTACKS:
-- Non-constant-time string comparison for tokens/secrets
-- Early return on auth failure leaking valid usernames
-
-Output JSON: [{id: "AUTH-XXX", ...full format with DREAD}]
-```
-
-### Agent 3: API Security Scanner (API)
-```
-Scan API endpoints for security issues:
-
-INPUT VALIDATION:
-- Missing input validation on request body
-- Type coercion attacks (string vs number)
-- Array/object pollution
-- Prototype pollution
-- Mass assignment vulnerabilities
-- GraphQL introspection enabled in production
-- GraphQL depth/complexity limits missing
-
-RATE LIMITING:
-- No rate limiting on expensive operations
-- Rate limit bypass via headers (X-Forwarded-For)
-- Non-atomic INCR+EXPIRE in Redis
-
-CORS MISCONFIGURATION:
-- res.header('Access-Control-Allow-Origin', req.headers.origin)
-- res.header('Access-Control-Allow-Origin', '*')
-- app.use(cors({ origin: true }))
-
-FAIL-OPEN vs FAIL-CLOSED PATTERNS (CRITICAL):
-- IP whitelist empty/missing = allow all
-- Auth middleware errors = request passes through
-- Rate limiter Redis down = no limiting
-- Config missing = insecure defaults
-
-WEBHOOKS:
-- Webhook signature not verified
-- SSRF via webhook URLs
-- No webhook replay protection
-
-Output JSON: [{id: "API-XXX", ...full format with DREAD}]
-```
-
-### Agent 4: Infrastructure Scanner (INFRA)
-```
-Scan infrastructure and deployment configs:
-
-DOCKER:
-- Running as root user
-- Secrets in Dockerfile or build args
-- Latest tag usage (unpinned versions)
-- Sensitive ports exposed
-- Missing health checks
-- No resource limits
-- Privileged mode enabled
+## Phase 1: Launch 22 Agents (parallel, background)
 
-KUBERNETES/HELM:
-- No resource limits/requests
-- Running as root
-- Privileged containers
-- Host network/PID enabled
-- Missing network policies
-- Secrets not encrypted at rest
-- Service account auto-mount enabled
-
-CI/CD:
-- Secrets in CI config files
-- Missing SAST/DAST in pipeline
-- No branch protection
-
-SECRETS IN GIT HISTORY (CRITICAL CHECK!):
-Run: git log --all --full-history -- "**/secrets*" "**/credentials*" "**/*.env"
-If secrets appear in history, they are EXPOSED even if now gitignored!
+**CRITICAL**: Each agent MUST:
+1. Scan the codebase for its specific issues
+2. Write findings to `.coverme/agents/{PREFIX}.json`
+3. Return ONLY the word "done" or "skipped" - NOTHING ELSE
 
-DEPENDENCY SECURITY (HIGH if missing):
-- npm audit or yarn audit in CI pipeline
-- Dependabot/Renovate configuration
-- SBOM generation
+Launch ALL with `run_in_background: true`:
 
-Output JSON: [{id: "INFRA-XXX", ...full format with DREAD}]
+### Agent 1: SEC - Security Core
 ```
-
-### Agent 5: Data & Privacy Scanner (DATA)
+Scan for: SQL injection, XSS, command injection, SSTI, hardcoded secrets, weak crypto.
+Write to .coverme/agents/SEC.json:
+[{"id":"SEC-001","title":"...","severity":"critical|high|medium|low","file":"...","line":N,"description":"...","recommendation":"..."}]
+Return ONLY: "done" or "skipped"
 ```
-Scan for data protection and privacy issues:
-
-PII HANDLING:
-- PII logged (emails, IPs, names, phone numbers)
-- PII in URLs/query strings
-- PII in error messages
-- PII not encrypted at rest
-- PII not masked in UI/logs
-
-GDPR/PRIVACY:
-- Missing data retention policy implementation
-- No data deletion mechanism (right to erasure)
-- No data export mechanism (data portability)
-- Consent not tracked properly
-- Third-party data sharing without consent
-
-DATABASE:
-- Sensitive data not encrypted (column-level)
-- No audit logging for sensitive operations
-- Connection strings with credentials in code
-
-SECRETS:
-- API keys in code (check git ls-files!)
-- Secrets in environment files committed
-- .env files not in .gitignore
 
-Output JSON: [{id: "DATA-XXX", ...full format with DREAD}]
+### Agent 2: AUTH - Authentication
 ```
-
-### Agent 6: AI/LLM Security Scanner (AI)
+Scan for: JWT issues, session problems, OAuth flaws, weak passwords, missing MFA.
+Write to .coverme/agents/AUTH.json
+Return ONLY: "done" or "skipped"
 ```
-CONDITIONAL: First detect if this project uses AI/LLM:
-grep -r -l "openai\|anthropic\|langchain\|ollama\|huggingface\|llama\|gpt-\|claude\|gemini\|bedrock\|vertex" --include="*.ts" --include="*.js" --include="*.py" --include="*.json" .
-
-IF NO AI CODE FOUND: Output: {"skipped": true, "reason": "No AI/LLM code detected", "findings": []}
-
-IF AI CODE FOUND:
-PROMPT INJECTION:
-- User input directly in prompts without sanitization
-- System prompts exposed to users
-- No input length limits on prompts
-- Missing output validation from LLM
-- Jailbreak vulnerabilities
 
-LLM OUTPUT → CODE EXECUTION CHAINS:
-- LLM generates SQL that gets executed
-- LLM generates code that gets eval'd
-- LLM generates shell commands that get executed
-- LLM output used in template rendering (SSTI)
-
-SUPPLY CHAIN:
-- CDN imports without Subresource Integrity (SRI)
-- Unpinned AI model versions
-
-Output JSON: [{id: "AI-XXX", ...full format with DREAD}]
+### Agent 3: API - API Security
 ```
-
-### Agent 7: Performance & DoS Scanner (PERF)
+Scan for: CORS issues, rate limiting, input validation, mass assignment, GraphQL issues.
+Write to .coverme/agents/API.json
+Return ONLY: "done" or "skipped"
 ```
-Scan for performance and denial-of-service issues:
 
-DATABASE:
-- N+1 query patterns
-- Missing indexes on filtered/sorted columns
-- Full table scans
-- Unbounded queries (no LIMIT)
-
-MEMORY:
-- Memory leaks (event listeners not removed)
-- Unbounded caches
-- Stream not properly closed
-- SSE/WebSocket buffer accumulation
-
-CPU:
-- ReDoS (Regular Expression DoS)
-- Algorithmic complexity attacks
-- Synchronous crypto operations
-- JSON parsing of large payloads
-
-DANGEROUS DATABASE OPERATIONS IN HOT PATHS:
-- Redis KEYS command (blocks entire server, O(n) scan)
-- MongoDB find() without limit
-- SQL SELECT without LIMIT
-
-Output JSON: [{id: "PERF-XXX", ...full format with DREAD}]
+### Agent 4: INFRA - Infrastructure
 ```
-
-### Agent 8: Business Logic Scanner (BIZ)
+Scan for: Docker issues, K8s misconfig, CI/CD secrets, cloud misconfig.
+Write to .coverme/agents/INFRA.json
+Return ONLY: "done" or "skipped"
 ```
-Scan for business logic vulnerabilities:
 
-RACE CONDITIONS:
-- TOCTOU (time-of-check-time-of-use)
-- Double-spend in transactions
-- Inventory overselling
-- Non-atomic read-modify-write
-
-WORKFLOW:
-- Step skipping in multi-step processes
-- State manipulation attacks
-- Workflow replay attacks
-
-FINANCIAL:
-- Rounding errors in calculations
-- Currency handling issues
-- Negative amount bypass
-- Discount stacking exploits
-
-Output JSON: [{id: "BIZ-XXX", ...full format with DREAD}]
+### Agent 5: DATA - Data & Privacy
 ```
-
-### Agent 9: Code Quality Scanner (QUAL)
+Scan for: PII exposure, GDPR issues, unencrypted data, secrets in code.
+Write to .coverme/agents/DATA.json
+Return ONLY: "done" or "skipped"
 ```
-Scan for code quality issues that affect security/reliability:
-
-COMPLEXITY:
-- Cyclomatic complexity > 10
-- Functions > 50 lines
-- Files > 500 lines
-- Deep nesting (> 4 levels)
 
-ANTI-PATTERNS:
-- God objects/classes
-- Callback hell
-- Magic numbers/strings
-- Dead code
-- Console.log in production
-- TODO/FIXME comments about security issues
-
-ERROR HANDLING:
-- Empty catch blocks
-- Generic error swallowing
-- Missing error boundaries (React)
-- Unhandled promise rejections
-
-DEAD CODE WITH SECURITY IMPLICATIONS:
-- Old/commented code that has BETTER security than current code
-- Deprecated functions with security controls not ported
-
-Output JSON: [{id: "QUAL-XXX", ...full format with DREAD}]
+### Agent 6: AI - AI/LLM Security
 ```
-
-### Agent 10: Testing & Reliability Scanner (TEST)
+FIRST: Check if AI code exists (openai, anthropic, langchain, etc.)
+If no AI code: Write {"skipped":true} and return "skipped"
+If AI code: Scan for prompt injection, data leakage, output validation.
+Write to .coverme/agents/AI.json
+Return ONLY: "done" or "skipped"
 ```
-Scan for testing gaps and reliability issues:
-
-TEST COVERAGE:
-- Critical paths without tests (auth, payments, data access)
-- Error handlers not tested
-- No integration tests
-- No E2E tests for main flows
 
-TEST QUALITY:
-- Tests without assertions
-- Mocked security checks (dangerous!)
-- Flaky tests (time-dependent)
-
-RELIABILITY:
-- Missing health checks
-- No graceful shutdown
-- Missing readiness/liveness probes
-- No circuit breakers for external calls
-- Missing retry logic with backoff
-
-Output JSON: [{id: "TEST-XXX", ...full format with DREAD}]
+### Agent 7: PERF - Performance & DoS
 ```
-
-### Agent 11: Redis & Cache Security Scanner (REDIS)
+Scan for: ReDoS, N+1 queries, memory leaks, unbounded operations.
+Write to .coverme/agents/PERF.json
+Return ONLY: "done" or "skipped"
 ```
-CONDITIONAL: First detect if this project uses Redis/Cache:
-grep -r -l "redis\|ioredis\|memcached\|node-cache\|lru-cache\|cache-manager" --include="*.ts" --include="*.js" --include="*.json" --include="*.yaml" .
-
-IF NO CACHE CODE FOUND: Output: {"skipped": true, "reason": "No Redis/Cache code detected", "findings": []}
-
-IF CACHE CODE FOUND:
-DANGEROUS COMMANDS:
-- KEYS * in production code (blocks entire server, use SCAN instead)
-- FLUSHALL, FLUSHDB accessible without protection
-- DEBUG, CONFIG commands enabled in production
-- EVAL/EVALSHA with user-controlled scripts (Lua injection)
-
-AUTHENTICATION & ACCESS:
-- Redis without AUTH (requirepass not set)
-- Redis exposed on 0.0.0.0 instead of 127.0.0.1
-- Missing TLS for Redis connections
-- Connection strings with passwords in code/logs
 
-DATA SECURITY:
-- Sensitive data stored without encryption
-- PII in Redis without TTL
-- Cache keys predictable/enumerable
-- No key prefix separation between tenants (multi-tenant leak)
-
-RACE CONDITIONS:
-- Non-atomic read-modify-write patterns
-- Missing WATCH/MULTI/EXEC for transactions
-- INCR + EXPIRE not atomic
-
-Output JSON: [{id: "REDIS-XXX", ...full format with DREAD}]
+### Agent 8: BIZ - Business Logic
 ```
-
-### Agent 12: Resilience & Fallback Scanner (RESIL)
+Scan for: Race conditions, TOCTOU, workflow bypass, financial issues.
+Write to .coverme/agents/BIZ.json
+Return ONLY: "done" or "skipped"
 ```
-Scan for resilience patterns and fallback mechanisms:
-
-CIRCUIT BREAKERS:
-- External service calls without circuit breaker
-- Circuit breaker without proper thresholds
-- No fallback when circuit is open
-
-RETRY PATTERNS:
-- Retries without exponential backoff
-- Retries without jitter (thundering herd)
-- Retries without max attempts limit
-- Retrying non-idempotent operations
 
-TIMEOUTS:
-- HTTP calls without timeout
-- Database queries without timeout
-- External API calls without timeout
-
-FALLBACKS:
-- No fallback for critical external dependencies
-- Missing cached fallback data
-- No degraded mode implementation
-
-HEALTH CHECKS:
-- Health check that calls external dependencies
-- No distinction between liveness and readiness
-
-Output JSON: [{id: "RESIL-XXX", ...full format with DREAD}]
+### Agent 9: QUAL - Code Quality
 ```
-
-### Agent 13: PII & Sensitive Data Scanner (PII)
+Scan for: High complexity, dead code, error swallowing, anti-patterns.
+Write to .coverme/agents/QUAL.json
+Return ONLY: "done" or "skipped"
 ```
-Scan for PII exposure and sensitive data handling issues:
-
-PII IN LOGS:
-- Email addresses logged
-- Phone numbers logged
-- IP addresses logged without justification
-- Names/addresses in logs
-- Request/response bodies logged without redaction
 
-PII IN URLS:
-- PII in URL path (e.g., /user/john@email.com)
-- PII in query parameters
-- Session tokens in URLs (referer leak)
-
-PII IN STORAGE:
-- PII stored without encryption at rest
-- PII in local storage/session storage (browser)
-- PII in cookies without encryption
-
-GDPR/CCPA COMPLIANCE:
-- No data export functionality
-- No data deletion on request
-- Missing consent tracking
-
-SENSITIVE DATA TYPES TO FIND:
-- SSN, passport numbers, national IDs
-- Credit card numbers (even partial)
-- Bank account numbers
-- Health/medical information
-- Authentication credentials
-
-Output JSON: [{id: "PII-XXX", ...full format with DREAD}]
+### Agent 10: TEST - Testing
 ```
-
-### Agent 14: Dead Code & Unused Scanner (DEAD)
+Scan for: Missing tests on critical paths, mocked security, no E2E.
+Write to .coverme/agents/TEST.json
+Return ONLY: "done" or "skipped"
 ```
-Scan for dead code, unused dependencies, and technical debt:
 
-UNUSED CODE:
-- Functions never called
-- Classes never instantiated
-- Variables assigned but never read
-- API endpoints not used by any client
-
-COMMENTED CODE:
-- Large blocks of commented-out code
-- TODO/FIXME comments older than 6 months
-- Console.log/print statements
-
-UNUSED DEPENDENCIES:
-- npm/pip packages installed but never imported
-- Devdependencies in production bundle
-- Duplicate dependencies (different versions)
-
-SECURITY IMPLICATIONS:
-- Commented security checks (why removed?)
-- Unused auth middleware (was security removed?)
-- Dead validation code (security regression?)
-
-Output JSON: [{id: "DEAD-XXX", ...full format with DREAD}]
+### Agent 11: REDIS - Cache Security
 ```
-
-### Agent 15: Database Security Scanner (DB)
+FIRST: Check if Redis/cache code exists.
+If not: Write {"skipped":true} and return "skipped"
+If yes: Scan for dangerous commands, auth issues, race conditions.
+Write to .coverme/agents/REDIS.json
+Return ONLY: "done" or "skipped"
 ```
-AUTO-DETECT DATABASE TYPE(S):
-| Database | Detection Pattern |
-|----------|-------------------|
-| PostgreSQL | pg, postgres, @prisma, typeorm, knex |
-| MySQL | mysql, mysql2, mariadb |
-| SQLite | sqlite, better-sqlite3, sql.js |
-| MongoDB | mongoose, mongodb |
-| DuckDB | duckdb, @duckdb |
-| Supabase | @supabase/supabase-js |
-
-IF NO DATABASE FOUND: Output: {"skipped": true, "reason": "No database code detected", "findings": []}
-
-SQL INJECTION:
-- String concatenation in queries
-- Template literals with user input in SQL
-- Raw queries without parameterization
-- Dynamic table/column names from user input
-
-NOSQL INJECTION:
-- MongoDB: $where with user input
-- MongoDB: operator injection ({$gt: ""})
-
-ACCESS CONTROL:
-- Database user with excessive privileges (GRANT ALL)
-- Application using root/admin database user
-- No row-level security (RLS) for multi-tenant
-
-CONNECTION SECURITY:
-- Database connection without TLS/SSL
-- Connection strings with credentials in code
-- Database exposed on public IP
 
-ORM SPECIFIC:
-- Prisma.$queryRawUnsafe() - VULNERABLE
-- Sequelize.query() with user input - VULNERABLE
-- TypeORM repository.query() with string concat - VULNERABLE
-- Mongoose User.find({ username: req.body.username }) - operator injection
-
-Output JSON: [{id: "DB-XXX", ...full format with DREAD}]
+### Agent 12: RESIL - Resilience
 ```
-
-### Agent 16: Network & Architecture Scanner (ARCH)
+Scan for: Missing circuit breakers, no timeouts, no retries, no fallbacks.
+Write to .coverme/agents/RESIL.json
+Return ONLY: "done" or "skipped"
 ```
-Scan for network architecture and service boundary issues:
-
-SERVICE BOUNDARIES:
-- Internal endpoints exposed externally
-- Missing network segmentation between services
-- Service-to-service communication without mTLS
-- Admin/debug ports accessible from outside
-
-For EACH FINDING, determine fixOwner:
-- Fixable by code? → fixOwner: "developer"
-- Fixable by NetworkPolicy/firewall/K8s config? → fixOwner: "devops"
-- Needs architecture redesign? → fixOwner: "architect"
-
-KUBERNETES/INFRASTRUCTURE:
-- Missing NetworkPolicies for namespace isolation
-- Services using ClusterIP that should be internal
-- LoadBalancer exposing internal services
-- Pod-to-pod communication without restrictions
 
-Output JSON: [{id: "ARCH-XXX", ...full format with notCodeFix: true if devops/architect}]
+### Agent 13: PII - PII Scanner
 ```
-
-### Agent 17: Design Decision Detector (DESIGN)
+Scan for: PII in logs, PII in URLs, unencrypted PII, missing GDPR controls.
+Write to .coverme/agents/PII.json
+Return ONLY: "done" or "skipped"
 ```
-Scan for intentional design decisions that might look like bugs.
-
-GOAL: Prevent false positives by identifying documented/intentional patterns
-
-DOCUMENTED DECISIONS:
-- Comments explaining WHY something is done a certain way
-- README/docs explaining architecture choices
-- ADR (Architecture Decision Records) files
-
-CODE PATTERNS THAT ARE NOT BUGS:
-- // Intentional: .... or // Design decision: ...
-- // SECURITY: This is safe because...
-- Feature flags controlling security features with documentation
-
-FOR EACH PATTERN FOUND, output:
-{
-  "id": "DESIGN-001",
-  "type": "documented_decision",
-  "title": "Content filtering disabled",
-  "file": "src/ai/chat.ts",
-  "line": 45,
-  "reason": "Documented in code comment",
-  "relatedFindings": ["AI-001"],
-  "recommendation": "Not a bug - document as accepted risk"
-}
 
-These findings will be EXCLUDED from main report and moved to "Design Decisions" section.
+### Agent 14: DEAD - Dead Code
 ```
-
-### Agent 18: Context-Aware Validator (CTX)
+Scan for: Unused functions, unused deps, commented code, TODO/FIXME.
+Write to .coverme/agents/DEAD.json
+Return ONLY: "done" or "skipped"
 ```
-Scan to understand the CONTEXT of each potential finding.
-
-GOAL: Reduce false positives by understanding deployment context
-
-FOR EACH FINDING FROM OTHER AGENTS, determine:
-
-DEPLOYMENT CONTEXT:
-- Is this code running in a container with network isolation?
-- Is this behind an API gateway that handles auth?
-- Is this internal-only service behind VPN?
 
-RUNTIME CONTEXT:
-- Is this code path actually reachable in production?
-- Is this only used in development/testing?
-- Is this protected by feature flag that's disabled?
-
-DATA FLOW CONTEXT:
-- Is the input already validated upstream?
-- Is there middleware that applies to this route?
-
-OUTPUT:
-{
-  "findingId": "SEC-001",
-  "contextAnalysis": {
-    "deploymentContext": "Runs in K8s with NetworkPolicy",
-    "runtimeContext": "Only reachable from internal services",
-    "verdict": "false_positive|confirmed|needs_review",
-    "reason": "Protected by network policy"
-  }
-}
+### Agent 15: DB - Database Security
 ```
-
-### Agent 19: Enclave & Trusted Compute Scanner (ENC)
+Scan for: SQL injection, NoSQL injection, missing RLS, exposed connections.
+Write to .coverme/agents/DB.json
+Return ONLY: "done" or "skipped"
 ```
-CONDITIONAL: First detect if this project uses enclaves/TEE:
-grep -r -l "enclave\|nitro\|sgx\|tee\|attestation\|trusted.compute\|confidential.computing\|vsock" --include="*.ts" --include="*.js" --include="*.py" --include="*.yaml" --include="*.yml" .
 
-IF NO ENCLAVE CODE FOUND: Output: {"skipped": true, "reason": "No enclave/TEE code detected", "findings": []}
-
-ENCLAVE REGISTRATION:
-- Enclave-to-backend registration without attestation
-- IP-based trust without cryptographic verification
-- Enclave secrets transmitted without encryption
-
-Output JSON: [{id: "ENC-XXX", ...full format with DREAD}]
+### Agent 16: ARCH - Architecture
 ```
-
-### Agent 20: Executive Summary Generator (EXEC)
+Scan for: Internal endpoints exposed, missing mTLS, network issues.
+Write to .coverme/agents/ARCH.json
+Return ONLY: "done" or "skipped"
 ```
-After all other agents complete, generate an executive summary.
 
-OUTPUT FORMAT:
-{
-  "executiveSummary": {
-    "headline": "3 Critical + 5 High findings require immediate attention",
-    "riskLevel": "HIGH",
-    "topRisks": [
-      "SQL injection in user search allows database access",
-      "Missing rate limiting enables brute force attacks"
-    ],
-    "positives": [
-      "Authentication flow is well-implemented",
-      "Good use of parameterized queries in core modules"
-    ],
-    "recommendedActions": [
-      {
-        "priority": 1,
-        "action": "Fix SQL injection in src/search.ts",
-        "owner": "developer",
-        "effort": "1-2 hours"
-      }
-    ],
-    "byOwner": {
-      "developer": 5,
-      "devops": 3,
-      "architect": 1
-    }
-  }
-}
+### Agent 17: DESIGN - Design Decisions
 ```
-
-### Agent 21: Duplicate & Existing Solutions Scanner (DUP)
+Find documented design decisions that might look like bugs (intentional patterns).
+Write to .coverme/agents/DESIGN.json
+Return ONLY: "done" or "skipped"
 ```
-CRITICAL: Before recommending ANY fix, check if a solution ALREADY EXISTS.
-
-For EVERY finding from Phase 1, search the codebase for:
-
-1. Existing utilities/helpers:
-   - Search for similar function names (sanitize, validate, escape)
-   - Check utils/, helpers/, lib/, common/ folders
-
-2. Existing patterns:
-   - How do OTHER files handle the same issue?
-   - Is there a project-wide convention?
-
-3. Duplicate findings:
-   - Is this the same issue reported multiple times?
-   - Are multiple findings actually ONE root cause?
 
-For EACH finding, add:
-- "EXISTING: Found sanitizeHtml() in src/utils/security.ts - use this"
-- "PATTERN: Other files use zod.string().email() - follow same pattern"
-- "DUPLICATE: Same root cause as SEC-003, fix once in middleware"
-
-Output:
-{
-  "findingId": "SEC-001",
-  "existingSolution": "Found: src/utils/sanitize.ts exports sanitizeUserInput()",
-  "duplicateOf": null,
-  "suggestedApproach": "Import and use existing function"
-}
+### Agent 18: CTX - Context Validator
 ```
-
-### Agent 22: Runtime Verification Scanner (RUNTIME)
+For critical findings from other agents, check deployment context.
+Write to .coverme/agents/CTX.json
+Return ONLY: "done" or "skipped"
 ```
-CONDITIONAL: Only runs if SSH is configured in Phase 0 runtime.json.
-
-IF NO runtime.json → Skip this agent entirely
-
-PURPOSE: Find dangerous mismatches between code configuration and actual runtime.
 
-STEP 1: Gather Code Expectations
-grep -E "^USER|^EXPOSE|^ENV" Dockerfile 2>/dev/null
-grep -E "runAsUser|runAsNonRoot|readOnlyRootFilesystem" k8s/*.yaml 2>/dev/null
-
-STEP 2: SSH and Check Actual Runtime
-ssh user@server "docker ps -q | head -1 | xargs -I {} docker exec {} id"
-ssh user@server "kubectl get pods -o name | head -1 | xargs -I {} kubectl exec {} -- id"
-
-STEP 3: Compare and Generate Findings
-
-| Mismatch | Severity |
-|----------|----------|
-| Code says non-root, runs as root | CRITICAL |
-| Dockerfile USER ignored | CRITICAL |
-| ReadOnlyRootFilesystem not enforced | HIGH |
-| Unexpected ports exposed | MEDIUM |
-
-Output JSON: [{id: "RUNTIME-XXX", expected: {...}, actual: {...}, ...full format}]
+### Agent 19: ENC - Enclave Security
 ```
-
----
-
-## Phase 2: Mitigation Validation (CRITICAL)
-
-Wait for all Phase 1 background agents to complete using `AgentOutputTool`.
-
-### Validator M: Mitigation Hunter
-
-For EVERY finding, search the codebase for mitigations:
-
-1. **Input Validation** - Is input sanitized before reaching the vulnerable sink?
-2. **Middleware Protection** - Is there middleware that protects this route?
-3. **Framework Auto-Protection** - Does the framework handle this automatically?
-4. **Route Protection** - Is the endpoint protected by auth/authorization?
-
-Output Per Finding:
-```json
-{
-  "findingId": "SEC-001",
-  "verdict": "mitigated|partial|confirmed|false_positive",
-  "mitigationType": "input_validation|middleware|framework|config|none",
-  "evidence": ["Found Zod validation at src/routes/user.ts:23"],
-  "adjustedSeverity": "low"
-}
+FIRST: Check if enclave/TEE code exists.
+If not: Write {"skipped":true} and return "skipped"
+If yes: Scan for attestation issues.
+Write to .coverme/agents/ENC.json
+Return ONLY: "done" or "skipped"
 ```
 
----
-
-## Phase 3: Cross-Validation (3 parallel validators)
-
-Launch 3 validators IN PARALLEL with `run_in_background: true`:
-
-### Validator A: False Positive Hunter
-Review ALL findings. For each finding:
-1. Read the actual code file
-2. Check if there are mitigating controls elsewhere
-3. For secrets: run "git ls-files <file>" - if not tracked, mark FALSE POSITIVE
-4. Check if code is actually reachable in production
-
-Output: { confirmed: ["SEC-001",...], falsePositives: [{id, reason},...] }
-
-### Validator B: Evidence Challenger
-Challenge every HIGH and CRITICAL finding:
-1. Read the actual code with 20 lines of context
-2. Trace data flow from source to sink
-3. Check for sanitization/validation in between
-4. Verify the exploit scenario is realistic
-
-Output: { confirmed: ["SEC-001",...], falsePositives: [{id, reason},...] }
+### Agent 20: EXEC - Executive Summary
+```
+After scanning, generate executive summary with top risks and positives.
+Write to .coverme/agents/EXEC.json
+Return ONLY: "done"
+```
 
-### Validator C: Missing Issues Hunter
-Look for issues that Phase 1 agents MISSED:
-- Race conditions in critical operations
-- Business logic flaws specific to this application
-- Combination attacks (multiple low issues = high)
+### Agent 21: DUP - Duplicate Finder
+```
+Find existing solutions in codebase that could fix other findings.
+Write to .coverme/agents/DUP.json
+Return ONLY: "done" or "skipped"
+```
 
-Output: { missedIssues: [{full finding object},...] }
+### Agent 22: POSITIVE - Good Patterns
+```
+Find positive security patterns and good practices in the codebase.
+Write to .coverme/agents/POSITIVE.json
+Return ONLY: "done"
+```
 
 ---
 
-## Phase 4: Build Consensus
-
-Wait for all Phase 3 background validators to complete using `AgentOutputTool`.
+## Phase 2: Wait for Agents
 
-Combine all results:
-1. Calculate confidence: (confirmations / validators) * 100
-2. Remove findings with confidence < 50%
-3. Add missed issues from Validator C
-4. Remove mitigated and false positive findings
-5. Identify positive observations (good patterns found)
+Wait for ALL background agents using `AgentOutputTool`.
+Each should return only "done" or "skipped".
 
 ---
 
-## Phase 5: Generate Report
-
-**DO NOT ASK - JUST OVERWRITE THE FILE!**
-
-Update `.coverme/scan.json` with the full scan results including:
+## Phase 3: Aggregate Results
 
-### Required Fields:
-- **projectName**: from package.json or folder name
-- **scanDate**: today's date
-- **filesScanned**: count of source files analyzed
-- **linesOfCode**: total lines in source files
-- **projectTree**: ASCII tree of main directories
-
-### projectOverview:
-```json
-{
-  "name": "project-name",
-  "type": "Backend API | Frontend SPA | Full-stack | CLI | Library",
-  "stack": ["Node.js", "TypeScript", "React", "PostgreSQL"],
-  "purpose": "Brief description of what this project does",
-  "architecture": "Monolith | Microservices | Serverless",
-  "keyComponents": ["auth service", "payment processing", "AI chat"]
-}
-```
+Read all agent files and merge:
 
-### architectureOverview:
-```json
-{
-  "components": [
-    {"name": "API Server", "type": "service", "description": "Express.js REST API", "trustLevel": "semi-trusted"},
-    {"name": "PostgreSQL", "type": "database", "description": "Primary data store", "trustLevel": "trusted"}
-  ],
-  "trustBoundaries": [
-    {"name": "Client to API", "from": "Browser", "to": "API Server", "protocol": "HTTPS"}
-  ],
-  "criticalAssets": [
-    {"name": "User Credentials", "type": "credential", "location": "PostgreSQL", "protection": "bcrypt hashed"}
-  ],
-  "dataFlows": ["User Authentication Flow", "Payment Processing Flow"]
-}
-```
-
-### executiveSummary:
-```json
-{
-  "headline": "3 Critical + 5 High findings require immediate attention",
-  "riskLevel": "CRITICAL | HIGH | MEDIUM | LOW",
-  "topRisks": ["SQL injection in user search", "Missing rate limiting"],
-  "positives": ["Good authentication flow", "Proper input validation"],
-  "recommendedActions": [
-    {"priority": 1, "action": "Fix SQL injection", "owner": "developer", "effort": "1-2 hours"}
-  ],
-  "byOwner": {"developer": 5, "devops": 3, "architect": 1}
-}
+```bash
+echo "Aggregating results..."
+ls -la .coverme/agents/
 ```
 
-### threatModel:
-```json
-[
-  {
-    "id": "T-001",
-    "threat": "SQL Injection via user input",
-    "category": "STRIDE",
-    "strideType": "T",
-    "status": "open | partial | mitigated",
-    "relatedFindings": ["SEC-001", "DB-002"],
-    "mitigation": "Use parameterized queries",
-    "dreadScore": 8.5
-  }
-]
-```
-
-### qualityReview:
-```json
-{
-  "deleteItems": [
-    {"type": "delete", "file": "src/utils/oldHelpers.ts", "lines": 250, "description": "Dead code", "reason": "No imports found"}
-  ],
-  "mergeItems": [
-    {"type": "merge", "file": "src/utils/validate.ts, src/helpers/validation.ts", "description": "Overlapping functions", "reason": "DRY violation"}
-  ],
-  "simplifyItems": [
-    {"type": "simplify", "file": "src/services/payment.ts", "lines": 450, "description": "Overly complex", "reason": "Can be reduced"}
-  ],
-  "totalLinesRemovable": 680,
-  "percentageOfCodebase": 4.2
-}
-```
+Use the Read tool to read each `.coverme/agents/*.json` file.
 
-### actionItems:
-```json
-[
-  {
-    "id": "A-001",
-    "title": "Fix SQL injection in user search",
-    "priority": "immediate | high | medium | long-term",
-    "relatedFindings": ["SEC-001"],
-    "effort": "low | medium | high",
-    "description": "Replace string concatenation with parameterized query"
-  }
-]
-```
-
-### findings:
-Each finding MUST include ALL these fields:
+Merge all findings into `.coverme/scan.json`:
 ```json
 {
-  "id": "SEC-001",
-  "title": "SQL Injection in User Search",
-  "severity": "critical | high | medium | low | info",
-  "category": "security | auth | api | infra | data | ai | performance | business-logic | quality | testing",
-  "file": "src/routes/users.ts",
-  "line": 45,
-  "code": "db.query(`SELECT * FROM users WHERE name LIKE '%${searchTerm}%'`)",
-  "description": "User search parameter directly concatenated into SQL query",
-  "impact": "Attacker can extract entire database contents",
-  "attackChain": [
-    {"step": 1, "action": "Send malicious search term", "result": "Query returns all users"},
-    {"step": 2, "action": "Use UNION SELECT", "result": "Database schema exposed"}
-  ],
-  "recommendation": "Use parameterized queries",
-  "cwe": "CWE-89",
-  "owasp": "A03:2021",
-  "dread": {
-    "damage": 10,
-    "reproducibility": 10,
-    "exploitability": 9,
-    "affectedUsers": 10,
-    "discoverability": 8,
-    "total": 9.4
-  },
-  "status": "open",
-  "confidence": 95,
-  "fixOwner": "developer | devops | architect",
-  "fixType": "code | config | infrastructure | design"
+  "projectName": "...",
+  "scanDate": "...",
+  "filesScanned": N,
+  "linesOfCode": N,
+  "findings": [...all from agent files...],
+  "positiveObservations": [...from POSITIVE.json...],
+  "summary": {"critical":N,"high":N,"medium":N,"low":N}
 }
 ```
 
-### Other sections:
-- **designDecisions**: Intentional patterns that look like bugs
-- **mitigatedFindings**: Findings with existing protection
-- **falsePositives**: Findings that are not actually vulnerabilities
-- **positiveObservations**: Good practices found
-- **validationSummary**: Stats on mitigation validation
-- **summary**: Total counts by severity
-- **agentsUsed**: List of agents that ran
-- **scanDuration**: Time taken in ms
-
-Use the Write tool to overwrite `.coverme/scan.json` with the results.
-
 ---
 
-## Phase 6: Generate HTML Report
-
-**DO NOT ASK - JUST RUN THE COMMANDS!**
+## Phase 4: Generate Report
 
-Generate the HTML report and open it automatically:
 ```bash
 TIMESTAMP=$(date +%Y-%m-%d_%H-%M-%S)
 npx coverme-scanner report .coverme/scan.json -f html -o ".coverme/report_$TIMESTAMP.html"
-cp .coverme/scan.json ".coverme/scan_$TIMESTAMP.json"
 open ".coverme/report_$TIMESTAMP.html"
 ```
 
-Run these commands without asking for permission.
-
 ---
 
 ## DONE
 
-Tell the user: "Scan complete! Report saved to .coverme/ and opened in browser."
-
-**REMINDER: You should have completed all 6 phases without asking ANY questions or stopping for confirmation.**
+Tell user: "Scan complete! Report opened."

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "coverme-scanner",
-  "version": "1.5.1",
+  "version": "1.6.0",
   "description": "AI-powered code scanner with multi-agent verification for Claude Code. One command scans everything.",
   "main": "dist/index.js",
   "files": [