Menu
SkillsMP has collected 146 skills from EntroVyx/hermes-agent-offsec. Open a skill to review its source and details.
Showing 40 of 146 collected skills.
Project self-knowledge for Hermes Agent Offsec. Use when the user asks about this fork, its CLI, configuration, update flow, providers, skills, modes, troubleshooting, or how it differs from the original Hermes Agent.
WordPress XML-RPC triage: method enum, multicall abuse, pingback SSRF, and upload-chain validation.
Complete bug bounty workflow — recon (subdomain enumeration, asset discovery, fingerprinting, HackerOne scope, source code audit), pre-hunt learning, vuln hunting (30+ classes), A-to-B chaining, AI/LLM testing, bypass tables, language-specific grep,…
Master bug bounty orchestrator — full pipeline: recon, pre-hunt learning, vulnerability hunting (30+ classes), A-to-B chaining, AI/LLM testing (ASI01-ASI10), language-specific grep, bypass tables, and reporting (7-question gate, CVSS 3.1, human-tone…
Join a Google Meet call, transcribe live captions, optionally speak in realtime, and do the followup work afterwards. Use when the user asks the agent to sit in on a meeting, take notes, summarize, respond in-call, or action items from it.
Attack SAML SSO via XSW, signature strip, metadata extract.
Chain multiple vulns into critical impact attack paths.
Execute optimal kill chains for WordPress full compromise.
Exploit Flask/Werkzeug debugger exposure — traceback disclosure, SECRET extraction, console RCE (if enabled), server path disclosure, and Python stack trace information gathering
Escape Docker containers to host root via 5 techniques.
Public-artifact-first methodology for turning exposed client assets, archived content, public API material, support content, and metadata into high-signal attack-surface maps. Use when starting recon, when a target looks thin from the homepage, or when you…
Visible exploit-chain builder. Use when a primitive appears and the operator needs to convert it into a reportable high-impact chain.
Catalog: 25 attacks, 18 WP, 8 CORS to match findings.
Client-runtime and bundle analysis methodology for converting front-end assets into route maps, auth assumptions, object graphs, and manual test queues. Use when JavaScript, source maps, or mobile-facing assets are available.
Methodology for turning exposed cloud, identity, storage, and backend-service clues into a structured capability assessment and manual validation plan. Use when recon reveals hosted data services, identity platforms, storage endpoints, or public backend…
Continuous recon and campaign-memory methodology for tracking deltas in attack surface, deduplicating noisy signals, and turning change events into prioritized manual testing queues. Use during long hunts or repeated target coverage.
Compare recon waves to find NEW, REGRESSED, PERSISTENT findings.
Attack-mode entry point for crown-jewel hunting. Use at the beginning of an authorized bug bounty or pentest session to choose the highest-impact objective before recon or exploitation.
Manual workflow attack mode for authorized bug bounty and pentest work. Use as /deep-hunt <workflow> to attack one business workflow deeply instead of scanning endpoints shallowly.
Offensive chain builder. Use when you have one or more weak findings and want to score, prioritize, and compose them into a critical or high-impact exploit path.
High-precision Google dorks for exposed configs, secrets, and credentials -- real-world validated
Manual testing trees for difficult bug bounty and pentest work. Use after mapping a workflow and before freeform poking so the agent systematically explores actor, state, transport, and secondary-object branches.
Master workflow for offensive campaigns. Use when starting a target, scoping a hunt, or deciding how to split work across reconnaissance, manual testing, chain building, and reporting.
7-phase pentest pipeline from passive recon to exploitation.
Service-first infrastructure triage methodology for turning discovered ports and banners into attack-surface meaning, weak-boundary hypotheses, and follow-up validation plans. Use when network services, certificates, reverse DNS, or exposed management…
4-phase pipeline for max findings per minute across batches.
Import HAR exports or raw Burp/Caido HTTP requests into an offensive request map. Use as /request-import <path> to extract endpoints, methods, parameters, cookies, auth hints, and deep-hunt candidates.
Pick sectors, compile targets, batch recon for campaigns.
How to split offensive campaigns across specialist subagents without duplicating work. Use when a target is large enough that one linear agent loop wastes time.
Create and maintain a persistent target dossier for authorized offsec campaigns. Use as /target-dossier <target> to initialize or update .offsec/campaigns/<target>/TARGET.md.
Persistent target-memory workflow for offsec campaigns. Use when you need structured memory for actors, objects, state transitions, secrets, hypotheses, and chain candidates on a real target.
Exploit no-auth APIs for data theft and CRUD via probes.
Poison CDN cache or deceive when X-Cache header is detected.
Sector-specific recon for carpet cleaning company websites — steam cleaning, stain removal, upholstery cleaning, flood restoration. Typically WordPress on shared hosting with service booking forms, emergency water damage response, and seasonal special offers.…
Exploit WP CORS credential reflection for data theft.
Deep pentest WP: SSRF, plugin CVE, JS mine, port scan chain.
DMARC/SPF/DKIM check, email spoofing, SMTP test, and security header analysis
Mine error_log for creds, paths, SQL when leak hunt finds.
Exchange/OWA NTLM AD leak, spray attack when mail subdomain.
Exploit Firebase/Supabase for data via JS config leak probe.