Skip to main content

contentdiscovery

Master content & URL discovery skill for exhaustive 50-point directory/file fuzzing, sensitive data leak auditing (.env, .git, backups), administrative console probing, and API endpoint extraction using TBHM v4 and tailored wordlists.

Source facts

Repository
pruthuraut/BB-skill
Last source activity
October 5, 2026 at 17:31
Detected SKILL.md language
English
Stars
0
Forks
0

Install options

The review-first prompt is selected by default. You can switch to a direct command or download a local copy.

Review the source files

Read SKILL.md and any companion files shown by SkillsMP before deciding whether to install.

File Explorer
11 files

Showing SKILL.md

SKILL.md
Source instructions · Read-only preview
name
contentdiscovery
description
Master content & URL discovery skill for exhaustive 50-point directory/file fuzzing, sensitive data leak auditing (.env, .git, backups), administrative console probing, and API endpoint extraction using TBHM v4 and tailored wordlists.
version
1.0.0
author
Bug Bounty Multi-Agent Skills Framework
compatibility
["Antigravity / Gemini CLI","Claude Code","Cursor IDE","OpenAI Codex / OpenCode","DeepSeek"]
# `contentdiscovery` — Universal Content & URL Discovery Skill ## Overview `contentdiscovery` is an offensive content enumeration and endpoint discovery skill designed to execute an exhaustive **50-point content discovery audit**. It implements Jason Haddix's *The Bug Hunter's Methodology (TBHM v4)*, integrating smart recursive fuzzing (`ffuf`, `feroxbuster`, `dirsearch`), historical URL mining (`gau`, `waybackurls`), JavaScript endpoint extraction (`katana`), sensitive file auditing (`.env`, `.git`, backups, database dumps), and enterprise/DevOps portal probing using tailored wordlists from `wordlists/` and `wordlists/`. ``` [ TARGET HOST / URL ] │ ┌─────────────────────────────┼─────────────────────────────┐ ▼ ▼ ▼ ┌─────────────────┐ ┌─────────────────┐ ┌─────────────────┐ │ Subagent 01 │ │ Subagent 02 │ │ Subagent 03 │ │ Fuzzers & │ │ Historical │ │ Well-Known, │ │ Crawlers │ │ JS & OSINT │ │ SEO & Identity │ │ (4 Checks) │ │ (5 Checks) │ │ (8 Checks) │ └────────┬────────┘ └────────┬────────┘ └────────┬────────┘ │ │ │ └─────────────────────────────┼─────────────────────────────┘ ▼ ┌─────────────────────────────┼─────────────────────────────┐ ▼ ▼ ▼ ┌─────────────────┐ ┌─────────────────┐ ┌─────────────────┐ │ Subagent 04 │ │ Subagent 05 │ │ Subagent 06 │ │ Sensitive Files│ │ Debug, Logs │ │ Admin Panels │ │ Backups & Git │ │ & API Docs │ │ & DevOps Cons. │ │ (12 Checks) │ │ (5 Checks) │ │ (16 Checks) │ └────────┬────────┘ └────────┬────────┘ └────────┬────────┘ │ │ │ └─────────────────────────────┴─────────────────────────────┘ ▼ [ FINAL CONSOLIDATED ARTIFACTS ] • discovered_endpoints.txt • sensitive_files.txt • exposed_admin_panels.txt • leaked_secrets.txt • checklist_50_tracker.md (50/50 Done) ``` --- ## The 6 Specialized Subagents 1. **[Subagent 01: High-Speed Recursive Fuzzers & Crawlers](file:///c:/Users/rautp/Documents/BBskill/.agents/skills/contentdiscovery/subagents/subagent_01_fuzzers_crawlers.md)** *Checks: 01, 02, 03, 04* *Wordlists:* `wordlists/MiniFuzz.txt`, `wordlists/God-Fuzz.txt`. 2. **[Subagent 02: Historical Crawling, JS Extraction & Public OSINT Leaks](file:///c:/Users/rautp/Documents/BBskill/.agents/skills/contentdiscovery/subagents/subagent_02_historical_js_osint.md)** *Checks: 22, 23, 24, 25, 26* *Methodology:* Archive mining (`gau`, `waybackurls`), live JS bundle crawling (`katana`), dev subdomain mapping, and Google Docs/Sheets dorking. 3. **[Subagent 03: Well-Known, SEO & Identity Federation Discovery](file:///c:/Users/rautp/Documents/BBskill/.agents/skills/contentdiscovery/subagents/subagent_03_wellknown_identity_seo.md)** *Checks: 05, 06, 07, 46, 47, 48, 49, 50* *Focus:* `robots.txt`, `sitemap.xml`, `security.txt`, OpenID Connect schemas, OAuth discovery, Android `assetlinks.json`, and iOS universal links. 4. **[Subagent 04: Sensitive Files, Source Leaks, Backups & Credentials](file:///c:/Users/rautp/Documents/BBskill/.agents/skills/contentdiscovery/subagents/subagent_04_sensitive_files_backups.md)** *Checks: 08, 09, 10, 11, 12, 13, 14, 27, 28, 29, 30, 31* *Wordlists:* `env.txt`, `dotfiles.txt`, `git_config.txt`, `zip.txt`, `sql.txt`, `config.txt`, `webconfig.txt`, `yaml.txt`. 5. **[Subagent 05: Debug Endpoints, Test Files & API Documentation](file:///c:/Users/rautp/Documents/BBskill/.agents/skills/contentdiscovery/subagents/subagent_05_debug_logs_api_docs.md)** *Checks: 15, 16, 17, 18, 34* *Wordlists:* `log.txt`, `phpunit.txt`, `api.txt`. 6. **[Subagent 06: Admin Consoles, DevOps Panels & Enterprise Portals](file:///c:/Users/rautp/Documents/BBskill/.agents/skills/contentdiscovery/subagents/subagent_06_admin_devops_portals.md)** *Checks: 19, 20, 21, 32, 33, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45* *Wordlists:* `phpmyadmin.txt`, `adminer.txt`, `WPfuzz.txt`, `JiraFuzz.txt`. For detailed wordlist categorization and real-world reproduction playbooks, see: - [Wordlists Guide & Resource Mapping](file:///c:/Users/rautp/Documents/BBskill/.agents/skills/contentdiscovery/references/wordlists_guide.md) - [CI/CD, DevOps Exposure & Information Disclosure Playbook](file:///c:/Users/rautp/Documents/BBskill/.agents/skills/contentdiscovery/references/cicd_devops_exposure_playbook.md) --- ## Master Orchestration Workflow When auditing a target (`TARGET_URL="https://example.com"`): ### Phase 1: Passive & Structural Harvesting (Subagents 02, 03) * Query archives via `waybackurls` and `gau`. * Extract endpoints from live JavaScript bundles (`katana -jc`). * Parse `robots.txt` Disallowed entries and `sitemap.xml`. * Audit `/.well-known/openid-configuration`, `security.txt`, and mobile link schemas. * Search for publicly indexed Google Spreadsheets/Docs. ### Phase 2: Active Recursive Fuzzing (Subagent 01) * Run `ffuf` with recursion depth 2 and smart status code filtering (`-mc 200,301,302,401,403 -ac`) using `MiniFuzz.txt`. * Run `dirsearch` with multiple extensions (`php,asp,aspx,jsp,html,json,txt`). ### Phase 3: Critical Exposures & Sensitive File Hunting (Subagent 04) * Fuzz for `.env` and `.env.local` using `wordlists/env.txt`. * Probe for exposed `.git/HEAD` and `.git/config` using `git_config.txt`. * Fuzz for database dumps (`.sql`, `.db`) using `sql.txt`. * Fuzz for backup archives (`.zip`, `.tar.gz`) using `zip.txt`. * Check for Dockerfiles, `package.json`, and `.htpasswd`. ### Phase 4: Debug Routines, Logs & Telemetry (Subagent 05) * Fuzz for log files (`error.log`, `access.log`) using `wordlists/log.txt`. * Probe for `/debug`, `/trace`, `/healthz`, and Prometheus `/metrics`. * Check for PHPUnit test files (`wordlists/phpunit.txt`) and `phpinfo.php`. * Fuzz for API documentation using `api.txt`. ### Phase 5: Administration Consoles & Middleware Portals (Subagent 06) * Fuzz for database managers (phpMyAdmin, Adminer). * Fuzz for CMS panels (WordPress `wp-admin`, Joomla). * Probe default DevOps ports: Grafana (`:3000`), Kibana (`:5601`), Airflow (`:8080`), RabbitMQ (`:15672`), Solr (`:8983`), MinIO (`:9001`). * Audit enterprise middleware consoles: Tomcat (`/manager/html`), JBoss, WebLogic, WebSphere. * Fuzz Atlassian Jira (`/secure/admin/`) and Confluence (`/admin/`) using `JiraFuzz.txt`. --- ## Standard Output Artifacts Each run populates the `artifacts/` folder: * `artifacts/discovered_endpoints.txt` — Deduplicated list of valid HTTP 200/301 endpoints. * `artifacts/sorted_categories/` — URL Categorization directory: - `api_urls.txt` (REST, GraphQL, Swagger, OpenAPI, JSON) - `admin_urls.txt` (Admin, Dashboard, Staff, Portals, Management) - `debug_urls.txt` (Debug, Actuator, Metrics, Health, Profiler, PHPinfo) - `db_urls.txt` (phpMyAdmin, Adminer, SQL dumps, database backups) - `php_urls.txt` (PHP scripts, legacy handlers, installation files) - `auth_urls.txt` (Login, Registration, 2FA, Password Reset, OAuth, SSO) - `sensitive_urls.txt` (Config files, .env, .git, backup archives) - `upload_urls.txt` (File upload, avatar, and attachment endpoints) - `idor_urls.txt` (Parameterized ID routes for IDOR / BOLA testing) - `params_urls.txt` (Parameterized query URLs for injection fuzzing) * `artifacts/sensitive_files.txt` — Discovered `.env`, `.git`, backup archives, and database dumps. * `artifacts/exposed_admin_panels.txt` — Reachable administrative consoles and DevOps dashboards. * `artifacts/checklist_50_tracker.md` — Complete 50/50 audit verification matrix.
View on GitHub