ZeeshanSultan/pentest-agent-vs-llm-benchmark-effectiveness

bounty-hunting-methodology

Bug bounty white-box hunting methodology. Load when the target is an open-source project with a security advisory program, bug bounty, or responsible disclosure policy.

analyst-overview

Root pointer for the analyst's vulnerability research playbooks. Load this first at iteration start to see the full catalog of vuln-class and chain-building skills.

trust-boundary-analysis

Trust boundary mapping and startup sequence audit for developer tools, CLI apps, and plugin systems. Load when the target is a developer tool, CLI, IDE extension, or any application that loads config from the current directory.

benchmark

CTF benchmark mode — meta-rules for automated benchmark runs. Routing/playbooks live in /skills/exploit/* and /skills/recon/*; this file only documents benchmark-specific conventions.

engagement-lifecycle

Red team engagement lifecycle management — initiation, phase transitions, go/no-go gates, deconfliction, emergency procedures, completion.

engagement-startup

Mandatory first-turn startup procedure — checks for existing engagements, resume/new selection, workspace initialization.

final-report

Final engagement report generation — executive summary, technical report, findings aggregation, attack path narrative, detection gap matrix, remediation roadmap.

orchestration

Decepticon orchestrator patterns — delegation, state management, adaptive re-planning, context handoff protocols.

ad

Active Directory exploitation — BloodHound analysis, Kerberoasting, AS-REP Roasting, AD CS abuse, DCSync, Golden Ticket, Constrained Delegation.

reporting

Exploitation finding documentation — initial access reports, exploit chain documentation, CVSS v4.0 scoring, shell/credential inventory, detection gap analysis.

web

Web application exploitation — the primary category skill for all web-based attacks. This is a routing skill: read this first to identify the attack type, then load the appropriate specialized sub-skill for detailed procedures. Covers 11 technique areas…

c2-sliver

Sliver C2 framework operations — server connection, listener setup, implant generation, BOF/Armory extensions, post-implant operations, HTTP C2 profiles.

reporting

Post-exploitation finding documentation — credential access, privilege escalation, lateral movement reports, detection gap analysis, attack path documentation, CVSS v4.0 scoring.

active-recon

Active target probing — port scanning, service detection, vulnerability scanning, banner grabbing, web directory fuzzing, SSL/TLS analysis.

cloud-recon

Cloud infrastructure enumeration — AWS S3 buckets, Azure blob storage, GCP buckets, cloud metadata endpoints, IAM misconfigurations, CDN origin detection.

osint

Open-source intelligence gathering — email harvesting, social media profiling, breach data checking, employee enumeration, GitHub secret scanning, organizational mapping.

passive-recon

Passive intelligence gathering without touching the target — DNS, WHOIS, subdomain enumeration, Certificate Transparency, technology fingerprinting, ASN mapping.

reporting

Recon output formatting — report structure, CVSS v4.0 scoring (primary), MITRE ATT&CK mapping, finding prioritization, Markdown output, detection gap tracking, handoff checklists.

web-recon

Web application enumeration hub — directory/file fuzzing, vhost discovery, API enumeration, CMS scanning, WAF detection, auth surface mapping, cookie audit.

finding-protocol

Finding documentation protocol — Markdown template, YAML frontmatter schema, CVSS v4.0 severity guide, confidence levels, naming conventions, post-creation checklist.

structured-questions

When and how to use ask_user_question — structured multiple-choice prompts vs. free-text prose during the engagement interview.

bounty-report-formatter

Bug bounty report formatting for HackerOne, Bugcrowd, Immunefi, and GitHub Security Advisories. Load after validate_finding succeeds and the finding needs to be submitted to a bounty program.

verifier-overview

Stage 3 triage and verification playbook. Crafts minimal PoCs, runs them with ZFP controls, promotes validated bugs to FINDING nodes with CVSS. Load at verifier-agent startup.