mirror of
https://github.com/x1xhlol/system-prompts-and-models-of-ai-tools.git
synced 2026-06-18 15:29:36 +00:00
3ef62652aa
Saves the DEALIX_PHASE2_EXECUTION_WAVES.md 90-day plan and scaffolds every
artifact the coding agent can produce. Wave A-E execution is explicitly
blocked until the Week-12 Phase Gate (§3) returns Green.
Added:
§1 Verification Protocol (V001-V007)
- scripts/v001_secret_scan.sh — trufflehog + gitleaks full-history scan
- backend/tests/security/test_rls_fuzz.py — 10K cross-tenant fuzz
- docs/verification/V003_pentest_engagement.md — vendor RFP + scope
- docs/verification/V004_no_founder_demo_test.md — 3-tester protocol
- scripts/v005_truth_registry_audit.py — independent audit tool
- infra/load-tests/baseline.js — k6 perf baseline
- frontend/tests/a11y/baseline.spec.ts — Playwright+axe baseline
- docs/baselines/README.md + docs/verification/README.md
§2 Founder Decision Sprint (FD001-FD005)
- docs/internal/legal_entity_decision.md — MISA/DIFC/Delaware brief
- docs/internal/trademark_status.md — SAIP filing kit tracker
- docs/hiring/{design_engineer, backend_engineer, head_of_cs}.md
§3 Customer Validation (CV001-CV004)
- docs/customer_learnings/pilot_agreement_template.md
- docs/customer_learnings/pilot_template/success_criteria.md
- docs/customer_learnings/pilot_template/kickoff_checklist.md
- docs/customer_learnings/friction_log.md + feature_requests.yaml
- docs/customer_learnings/weekly_review_template.md
Truth registry updates
- docs/registry/TRUTH.yaml — new verification_protocol,
founder_decision_sprint, customer_validation sections
Gates (post-change):
architecture_brief.py 40/40
release_readiness_matrix 94/94 (added 30 new scaffold checks)
v005_truth_registry_audit 19/19 SUPPORTED
100 lines
3.4 KiB
Markdown
100 lines
3.4 KiB
Markdown
# V003 — External Penetration Test Engagement
|
||
|
||
> **Status**: NOT STARTED — founder action required
|
||
> **Gate**: Phase 2 cannot claim "pentested" until written report exists in `docs/internal/pentest_report_YYYYMMDD.pdf`
|
||
> **Budget**: $20,000 – $40,000 USD
|
||
> **Target completion**: Week 10
|
||
|
||
---
|
||
|
||
## Vendor Shortlist
|
||
|
||
| Vendor | Strengths | Indicative Quote | Region | Link |
|
||
|--------|-----------|------------------|--------|------|
|
||
| **Cure53** | Browser + web app focus; strong LLM/prompt-injection experience | $25–35K | Berlin | https://cure53.de |
|
||
| **Trail of Bits** | Deep protocol + cryptography + supply chain | $35–50K | NYC | https://www.trailofbits.com |
|
||
| **NCC Group** | Enterprise-grade, global presence, SOC 2 alignment | $30–45K | London/NYC | https://www.nccgroup.com |
|
||
| **Securinc** | MENA-focused, Arabic+English reporting | $15–25K | Dubai | https://securinc.io |
|
||
| **Include Security** | Web + LLM + cloud posture | $25–40K | USA | https://includesecurity.com |
|
||
|
||
---
|
||
|
||
## Required Scope (send to vendors verbatim)
|
||
|
||
1. **Authentication & Session**
|
||
- JWT lifecycle, refresh token rotation, session fixation
|
||
- SSO/SCIM flows (once WorkOS in place — Wave B)
|
||
- MFA bypass attempts
|
||
|
||
2. **Multi-Tenancy Isolation**
|
||
- PostgreSQL Row-Level Security bypass attempts
|
||
- Cross-tenant data access via ORM, raw SQL, IDOR
|
||
- Tenant context tampering via JWT claims
|
||
|
||
3. **Authorization (ABAC)**
|
||
- Policy class A/B/C enforcement (Approval Bridge)
|
||
- Approval workflow forgery
|
||
- Evidence Pack tampering
|
||
|
||
4. **LLM & Prompt Injection**
|
||
- OWASP LLM Top 10 across all 17 structured output endpoints
|
||
- Prompt leakage (model_router, partner dossier, Saudi workflow)
|
||
- Jailbreak via Arabic/RTL encoding tricks
|
||
- Training data leakage via echo attacks
|
||
|
||
5. **File Uploads / Evidence**
|
||
- Path traversal on uploads
|
||
- Polyglot file attacks
|
||
- SHA256 tamper detection bypass
|
||
|
||
6. **Webhooks / Integrations**
|
||
- Signature forgery on WhatsApp/Email/ZATCA webhooks
|
||
- Replay attacks
|
||
- SSRF via outbound connectors
|
||
|
||
7. **Infrastructure**
|
||
- Container escape (if applicable)
|
||
- Redis command injection
|
||
- CORS / CSP review
|
||
|
||
---
|
||
|
||
## Deliverables (required from vendor)
|
||
|
||
1. Executive summary (1–2 pages, Arabic + English preferred)
|
||
2. Technical findings per OWASP risk rating (Critical / High / Medium / Low / Info)
|
||
3. Reproducer steps for every finding
|
||
4. Re-test report after remediation
|
||
5. Letter of attestation suitable for customer security questionnaires
|
||
|
||
---
|
||
|
||
## Acceptance Criteria (Day 90)
|
||
|
||
- [ ] Vendor engaged with SOW signed
|
||
- [ ] Report received (PDF or signed Markdown)
|
||
- [ ] 0 open Critical findings
|
||
- [ ] ≤2 open High findings (with remediation plan)
|
||
- [ ] Re-test scheduled
|
||
|
||
---
|
||
|
||
## Founder Checklist
|
||
|
||
- [ ] Shortlist 3 vendors from table above
|
||
- [ ] Send identical RFP; compare price + scope + timeline
|
||
- [ ] Legal: confirm NDA in place before sharing architecture docs
|
||
- [ ] Legal: confirm whether SAR or USD invoicing (KSA VAT implications)
|
||
- [ ] Allocate technical point-of-contact (founder or senior engineer)
|
||
- [ ] Schedule kickoff call with vendor
|
||
- [ ] Provide vendor: staging URL, test accounts (Tenant A, Tenant B, admin), architecture brief, this scope doc
|
||
|
||
---
|
||
|
||
## Anti-Patterns
|
||
|
||
- ❌ Claiming "pentested" based on automated scans (Snyk, Trivy, Burp alone)
|
||
- ❌ Claiming "pentested" based on internal red-team exercise
|
||
- ❌ Time-limited engagement <5 business days
|
||
- ❌ Accepting a vendor whose report template has <10 pages
|