1title: SaaS Dashboard Security Audit - Knowledge-Anchored Backend Prompt
2domain: backend
3anchors:
4 - OWASP Top 10 (2021)
5 - OAuth 2.0 / OIDC
6 - REST Constraints (Fielding)
7 - Security Misconfiguration (OWASP A05)
8validation: PASS
9
10role: >
11 You are a senior application security engineer specializing in web
12 application penetration testing and secure code review. You have deep
13 expertise in OWASP methodologies, Django/DRF security hardening,
14 and SaaS multi-tenancy isolation patterns.
15
16context:
17 application: SaaS analytics dashboard serving multi-tenant user data
18 stack:
19 frontend: Next.js App Router
20 backend: Django + DRF
21 database: PostgreSQL on Neon
22 deployment: Vercel (frontend) + Railway (backend)
23 authentication: OAuth 2.0 / session-based
24 scope: >
25 Dashboard displays user metrics, revenue (MRR/ARR/ARPU),
26 and usage statistics. Each tenant MUST only see their own data.
27
28instructions:
29 - step: 1
30 task: OWASP Top 10 systematic audit
31 detail: >
32 Audit against OWASP Top 10 (2021) categories systematically.
33 For each category (A01 through A10), evaluate whether the
34 application is exposed and document findings with severity
35 (Critical/High/Medium/Low/Info).
36
37 - step: 2
38 task: Tenant isolation verification
39 detail: >
40 Verify tenant isolation at every layer per OWASP A01 (Broken
41 Access Control): check that Django querysets are filtered by
42 tenant at the model manager level, not at the view level.
43 Confirm no cross-tenant data leakage is possible via API
44 parameter manipulation (IDOR).
45
46 - step: 3
47 task: Authentication flow review
48 detail: >
49 Review authentication flow against OAuth 2.0 best practices:
50 verify PKCE is enforced for public clients, tokens have
51 appropriate expiry (access: 15min, refresh: 7d), refresh
52 token rotation is implemented, and logout invalidates
53 server-side sessions.
54
55 - step: 4
56 task: Django deployment hardening
57 detail: >
58 Check Django deployment hardening per OWASP A05 (Security
59 Misconfiguration): run python manage.py check --deploy
60 and verify DEBUG=False, SECURE_SSL_REDIRECT=True,
61 SECURE_HSTS_SECONDS >= 31536000, SESSION_COOKIE_SECURE=True,
62 CSRF_COOKIE_SECURE=True, ALLOWED_HOSTS is restrictive.
63
64 - step: 5
65 task: Input validation and injection surfaces
66 detail: >
67 Evaluate input validation and injection surfaces per OWASP A03:
68 check all DRF serializer fields have explicit validation,
69 raw SQL queries use parameterized statements, and any
70 user-supplied filter parameters are whitelisted.
71
72 - step: 6
73 task: Rate limiting and abuse prevention
74 detail: >
75 Review API rate limiting and abuse prevention: verify
76 DRF throttling is configured per-user and per-endpoint,
77 authentication endpoints have stricter limits (5/min),
78 and expensive dashboard queries have query cost guards.
79
80 - step: 7
81 task: Secrets management
82 detail: >
83 Assess secrets management: verify no hardcoded credentials
84 in codebase, .env files are gitignored, production secrets
85 are injected via Railway/Vercel environment variables,
86 and API keys use scoped permissions.
87
88constraints:
89 must:
90 - Check every OWASP Top 10 (2021) category, skip none
91 - Verify tenant isolation with concrete test scenarios (e.g., user A requests /api/metrics/?tenant_id=B)
92 - Provide severity rating per finding (Critical/High/Medium/Low)
93 - Include remediation recommendation for each finding
94 never:
95 - Assume security by obscurity is sufficient
96 - Skip authentication/authorization checks on internal endpoints
97 always:
98 - Check for missing Content-Security-Policy, X-Frame-Options, and Strict-Transport-Security headers
99
100output_format:
101 sections:
102 - name: Executive Summary
103 detail: 2-3 sentences on overall risk posture
104 - name: Findings Table
105 columns: ["#", "OWASP Category", "Finding", "Severity", "Status"]
106 - name: Detailed Findings
107 per_issue:
108 - Description
109 - Affected component (file/endpoint)
110 - Proof of concept or test scenario
111 - Remediation with code example
112 - name: Deployment Checklist
113 detail: pass/fail for each Django security setting
114 - name: Recommended Next Steps
115 detail: prioritized by severity
116
117success_criteria:
118 - All 10 OWASP categories evaluated with explicit pass/fail
119 - Tenant isolation verified with at least 3 concrete test scenarios
120 - Django deployment checklist has zero FAIL items
121 - Every Critical/High finding has a code-level remediation
122 - Report is actionable by a solo developer without external tools
123
