Skip to main content

credentialattacks

Master authentication and credential security skill for exhaustive 40-point password policy auditing, brute-force & rate-limiting defenses, password reset flow analysis, and 2FA/MFA bypass testing based on TBHM v4.

معلومات المصدر

المستودع
pruthuraut/BB-skill
آخر نشاط في المصدر
٥ أكتوبر ٢٠٢٦ في ١٧:٣١
لغة SKILL.md المكتشفة
الإنجليزية
النجوم
٠
التفرعات
٠

خيارات التثبيت

يُحدَّد Prompt الذي يراجع المصدر أولًا بشكل افتراضي. يمكنك التبديل إلى أمر مباشر أو تنزيل نسخة محلية.

مراجعة ملفات المصدر

اقرأ SKILL.md وأي ملفات مرافقة يعرضها SkillsMP قبل أن تقرر التثبيت.

مستكشف الملفات
9 ملفات

عرض SKILL.md

SKILL.md
تعليمات المصدر · معاينة للقراءة فقط
name
credentialattacks
description
Master authentication and credential security skill for exhaustive 40-point password policy auditing, brute-force & rate-limiting defenses, password reset flow analysis, and 2FA/MFA bypass testing based on TBHM v4.
version
1.0.0
author
Bug Bounty Multi-Agent Skills Framework
compatibility
["Antigravity / Gemini CLI","Claude Code","Cursor IDE","OpenAI Codex / OpenCode","DeepSeek"]
# `credentialattacks` — Universal Password & Credential Security Skill ## Overview `credentialattacks` is a comprehensive offensive authentication security skill designed to execute an exhaustive **40-point credential security audit**. It audits password policy boundaries, brute-force and credential stuffing protections, IP rate-limit header bypasses (`X-Forwarded-For`), password reset flows (token entropy, Host header injection takeovers), and multi-factor authentication (2FA/MFA) bypasses (direct access, response manipulation, OTP brute forcing) based on Jason Haddix's *The Bug Hunter's Methodology (TBHM v4)* and real-world bounty notes. ``` [ TARGET AUTHENTICATION ] │ ┌────────────────────────────┼────────────────────────────┐ ▼ ▼ ▼ ┌─────────────────┐ ┌─────────────────┐ ┌─────────────────┐ │ Subagent 01 │ │ Subagent 02 │ │ Subagent 03 │ │ Password Policy│ │ Brute-Force & │ │ Reset Flows & │ │ & Input Flaws │ │ Rate Limiting │ │ Takeover Risks │ │ (8 Checks) │ │ (8 Checks) │ │ (7 Checks) │ └────────┬────────┘ └────────┬────────┘ └────────┬────────┘ │ │ │ └────────────────────────────┼────────────────────────────┘ ▼ ┌────────────────────────────┴────────────────────────────┐ ▼ ▼ ┌─────────────────┐ ┌─────────────────┐ │ Subagent 04 │ │ Subagent 05 │ │ 2FA/MFA Bypass │ │ Client-Side │ │ & OTP Auditing │ │ Transport & Sess│ │ (10 Checks) │ │ (7 Checks) │ └────────┬────────┘ └────────┬────────┘ │ │ └────────────────────────────┬────────────────────────────┘ ▼ [ CONSOLIDATED AUDIT ARTIFACTS ] • credential_audit_report.json • 2fa_bypasses.txt • rate_limit_bypasses.txt • account_takeover_risks.txt • checklist_40_tracker.md (40/40 Complete) ``` --- ## The 5 Specialized Subagents 1. **[Subagent 01: Password Policies, Encodings & Input Flaws](file:///c:/Users/rautp/Documents/BBskill/.agents/skills/credentialattacks/subagents/subagent_01_password_policies_input.md)** *Checks: 01, 07, 08, 14, 15, 16, 33, 34* *Focus:* Password complexity, old password verification, password reuse, bcrypt 72-byte truncation, Unicode normalization, SQL injection in password field, and hashing CPU DoS. 2. **[Subagent 02: Default Credentials, Brute Force & Rate Limiting Defenses](file:///c:/Users/rautp/Documents/BBskill/.agents/skills/credentialattacks/subagents/subagent_02_bruteforce_ratelimit_defaults.md)** *Checks: 02, 03, 09, 10, 11, 12, 35, 40* *Focus:* Default service credentials, credential stuffing velocity limits, `X-Forwarded-For` IP spoofing bypasses, account lockout user enumeration, password comparison timing attacks, and CAPTCHA bypasses. 3. **[Subagent 03: Password Reset Flows & Account Takeovers](file:///c:/Users/rautp/Documents/BBskill/.agents/skills/credentialattacks/subagents/subagent_03_reset_flows_takeovers.md)** *Checks: 04, 05, 06, 13, 31, 32, 36* *Focus:* Token PRNG entropy, GET parameter leakage, email enumeration, single-use token replay, Host header injection account takeovers, and plaintext password storage. 4. **[Subagent 04: Multi-Factor Authentication (2FA/MFA) Bypass Auditing](file:///c:/Users/rautp/Documents/BBskill/.agents/skills/credentialattacks/subagents/subagent_04_2fa_mfa_bypass.md)** *Checks: 20, 21, 22, 23, 24, 25, 26, 27, 28, 30* *Focus:* Direct endpoint navigation (`/dashboard`), post-auth API token access without 2FA claim, OTP brute force, sequential codes, code reuse race conditions, backup code entropy, unverified 2FA deactivation, and response status-code tampering (`400/403` to `200 OK`). 5. **[Subagent 05: Client-Side Security, Transport & Session Policies](file:///c:/Users/rautp/Documents/BBskill/.agents/skills/credentialattacks/subagents/subagent_05_client_transport_sessions.md)** *Checks: 17, 18, 19, 29, 37, 38, 39* *Focus:* HTTPS enforcement, browser autocomplete attributes, `localStorage` / `sessionStorage` credential leakage, step-up 2FA re-authentication on sensitive actions, and concurrent multi-session revocation. For rate-limit IP header bypasses and step-by-step reproduction guides, refer to: - [Rate Limiting & IP Header Spoofing Reference](file:///c:/Users/rautp/Documents/BBskill/.agents/skills/credentialattacks/references/rate_limit_bypass_headers.md) - [OTP Rate Limiting, Throttling Bypass & Financial Resource Exhaustion Playbook](file:///c:/Users/rautp/Documents/BBskill/.agents/skills/credentialattacks/references/otp_rate_limiting_and_dos_playbook.md) --- ## Master Orchestration Workflow When executing this skill against a target authentication system: ### Phase 1: Policy & Input Validation (Subagent 01) * Test password registration with 1-character, numeric, and trivial passwords. * Test password change omitting the current password parameter. * Submit a password longer than 72 bytes to test for bcrypt truncation. * Test Unicode normalization and SQL injection payloads (`' OR '1'='1`) in password fields. ### Phase 2: Brute-Force, Rate Limiting & Lockout Auditing (Subagent 02) * Check default administrative credentials (`admin:admin`, `tomcat:s3cret`). * Execute 30 rapid login attempts to measure velocity rate limiting. * Test header-based IP rotation (`X-Forwarded-For`, `X-Client-IP`, `X-Originating-IP`). * Measure account lockout thresholds and verify if lockout messages leak user existence. * Test CAPTCHA validation (omission, empty token, token replay). ### Phase 3: Password Reset Lifecycle & Account Takeover (Subagent 03) * Test password reset for user enumeration via status code, body, or timing deltas. * Collect multiple reset tokens to evaluate PRNG entropy and timestamp dependency. * Verify if reset tokens are sent in URL query strings without restrictive Referrer-Policy. * Test token single-use invalidation by replaying a completed reset token. * Inject `Host: evil.com` and `X-Forwarded-Host: evil.com` into password reset requests to test for reset poisoning. ### Phase 4: 2FA / MFA Multi-Factor Bypass Auditing (Subagent 04) * Complete Step 1 login; attempt direct navigation to `/dashboard` and call authenticated APIs. * Test 2FA OTP brute-forcing and velocity limits on `/verify-2fa`. * Intercept failed OTP responses and tamper HTTP status code from `400/403` to `200 OK`. * Test sending concurrent requests with the same valid OTP to verify one-time use. * Attempt disabling 2FA without entering the current OTP code. ### Phase 5: Client-Side Credential Storage & Transport Security (Subagent 05) * Verify strict HTTPS submission for login and reset forms. * Inspect `localStorage` and `sessionStorage` for cached credentials post-login. * Audit password input fields for proper `autocomplete` attributes (`current-password`, `new-password`). * Test step-up 2FA requirements on sensitive account actions (email/payout changes). --- ## Standard Output Artifacts Each run populates the `artifacts/` folder: * `artifacts/credential_audit_report.json` — Consolidated findings across all 40 checks. * `artifacts/2fa_bypasses.txt` — Confirmed multi-factor authentication bypass vectors. * `artifacts/rate_limit_bypasses.txt` — Successful IP-spoofing and velocity bypass headers. * `artifacts/account_takeover_risks.txt` — Password reset poisoning and token reuse findings. * `artifacts/checklist_40_tracker.md` — 40/40 completed verification matrix.
عرض على GitHub