| name | recon |
| description | Perform structured reconnaissance and attack surface enumeration for authorized penetration tests, CTF challenges, and bug bounty programs. Use when the user mentions 'recon,' 'reconnaissance,' 'enumerate,' 'attack surface,' 'subdomain enumeration,' 'port scan,' 'fingerprint,' 'asset discovery,' or needs to map a target's external footprint. |
| allowed-tools | Bash, Read, Write, WebSearch, WebFetch |
Recon — Penetration Testing Reconnaissance
Perform structured reconnaissance against an authorized target, organizing findings into an actionable attack surface map.
Cross-references: osint-recon for the deeper open-source-intelligence pass (people, organizations, historical data) — this skill is the active/passive target-mapping side, osint-recon is the broader investigative side; they pair naturally. web-pentest for the next stage once recon has produced an attack surface map and an authorized target list. owasp-audit for source-code review when you have access to the target's code.
Authorization Check
Before running any commands, confirm:
- The user has written authorization for the target (pentest engagement, bug bounty program, CTF/lab environment)
- The target is within the defined scope
If authorization is unclear, ask before proceeding. Never assume authorization.
Methodology
Phase 1: Passive Recon
Gather information without touching the target directly.
DNS enumeration:
WHOIS and registration: Run whois $ARGUMENTS for registrant, nameserver, and creation date info.
Search engine dorking: Use targeted queries — site:, inurl:, filetype:, intitle: — to find exposed pages, documents, and admin panels.
Technology fingerprinting: Identify frameworks, CMS, server software, and JavaScript libraries from public-facing pages.
Public code repositories: Search GitHub/GitLab for the target's org name, domain, API keys, or internal paths.
Historical data: Check the Wayback Machine for old endpoints, removed pages, and configuration files.
Phase 2: Active Recon (explicit authorization only)
Port scanning:
nmap -sC -sV -oN scan-results.txt $ARGUMENTS
Start with top 1000 ports. Expand to full range (-p-) if needed. Use -Pn if the host appears down but is in scope.
Service enumeration: Based on open ports, probe for version info and default configurations.
Web content discovery:
- Directory bruting with gobuster, feroxbuster, or dirsearch
- Virtual host enumeration
- API endpoint discovery (check
/api/, /v1/, /graphql, /swagger.json)
SSL/TLS analysis: Run testssl.sh or sslyze to check for weak ciphers, expired certificates, and misconfigurations.
Phase 3: Analysis
Correlate all findings. Identify the most promising attack vectors and prioritize by:
- Severity of potential impact
- Likelihood of exploitation
- Exposure level (internet-facing vs. internal)
Output Format
Produce a structured recon report:
# Recon Report
## Target: [target]
## Scope: [confirmed scope]
## Date: [date]
### Passive Findings
| Finding | Details | Relevance |
|---------|---------|-----------|
### Subdomains Discovered
- [list]
### Technologies Detected
- [list with versions where identified]
### Active Findings
| Port | Service | Version | Notes |
|------|---------|---------|-------|
### Attack Surface Summary
[Prioritized list of interesting findings with risk assessment]
### Recommended Next Steps
[Ordered list of what to investigate further]
Boundaries
- Stay within the defined scope — never scan adjacent or out-of-scope systems
- Rate-limit aggressive scans to avoid disruption
- Log all commands run for the engagement record
- If you discover evidence of active compromise by a third party, alert the user immediately
- Refuse requests targeting systems without explicit authorization
- Refuse requests for mass scanning of unrelated targets
References
- PTES (Penetration Testing Execution Standard)
- OWASP Testing Guide
- Bug Bounty Methodology (jhaddix/tbhm)
