Attack Chain Orchestration
Plans multi-stage authorized penetration tests and routes each stage to specialist skills.
- Repository
- zhaoxuya520/reverse-skill
- Stars
- 43,969
- Licence
- MIT
- Files
- 4 files
What it does
This skill turns the agent into a planner for multi-step offensive security work, such as moving from an external foothold to an internal domain controller. It lays out phases (reconnaissance, initial access, privilege escalation, lateral movement, impact) with concrete tool commands and attack techniques, and it tells the agent to hand single-stage tasks to other specialist skills. Its own instructions require an authorization and scope file to be created before any action, though the visible text is largely a reference catalogue of attack methods.
Great for
- Security professionals running an authorized penetration test or red-team exercise who want a structured attack-path plan
- Teams preparing a scoped assessment of an organisation's exposure across multiple stages
Skip it if
- Anyone without written authorization to test the target: the content is a catalogue of intrusion, phishing and credential-attack techniques, and misuse against systems you do not own is illegal and harmful
- Beginners looking for a general learning guide: it assumes working knowledge of tools like nmap, sqlmap and Active Directory attacks and does not explain safe lab practice
Try asking
- “Plan a full authorized external-to-internal penetration test for our company's staging network, with a scope document first”
- “I have a webshell on a test server; what are the realistic next stages and which specialist skill should handle each?”
- “Map the attack surface of this in-scope domain and propose the shortest attack path to the critical system”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 4 original files, the MIT licence and an attribution note.
attack-chain.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip attack-chain.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip attack-chain.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Provide a signed scope and authorization record before asking for any action, since the skill itself gates execution on auth status
- Use it for planning and sequencing, then run individual stages through the specialist skills it names rather than letting one agent run everything unattended
What's inside
- references/attack-playbooks.md
- references/evasion-cheatsheet.md
- references/lifecycle-checklist.md
- SKILL.md
Redistributed unmodified from zhaoxuya520/reverse-skill under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.