using-offensive-claude

Name: using-offensive-claude
Author: hypnguyen1209

320stars55forks

How to install using-offensive-claude?

hypnguyen1209/offensive-claude/using-offensive-claude

$npx -y skills add hypnguyen1209/offensive-claude --skill using-offensive-claude

Installs into the current project.

Summary

Use when starting any offensive-security engagement or task — establishes how to find and invoke the right skill before any action (including clarifying questions, recon, exploitation, or reporting)

Files · 1

View on GitHub

SKILL.md

1	<SUBAGENT-STOP>
2	If you were dispatched as a subagent to execute a specific task, skip this skill.
3	</SUBAGENT-STOP>
4
5	<EXTREMELY-IMPORTANT>
6	If there is even a 1% chance a skill applies to what you are doing, you ABSOLUTELY MUST invoke it.
7
8	IF A SKILL APPLIES TO YOUR TASK, YOU DO NOT HAVE A CHOICE. YOU MUST USE IT.
9
10	This is not negotiable. You cannot rationalize your way out of it.
11	</EXTREMELY-IMPORTANT>
12
13	# Using Offensive-Claude
14
15	You are operating an authorized offensive-security framework. Every action assumes a
16	prior, written authorization whose boundary is declared in `scope.json` (see scope-discipline).
17
18	## Instruction Priority
19
20	1. User's explicit instructions (CLAUDE.md, direct requests) — highest.
21	2. These skills — override default behavior where they conflict.
22	3. Default behavior — lowest.
23
24	User instructions say WHAT, not HOW. "Exploit X" or "scan Y" does not mean skip the
25	discipline skills (scope, finding, OPSEC). The one thing the operator cannot waive is the
26	authorization boundary — see scope-discipline.
27
28	## The Rule
29
30	Invoke relevant skills BEFORE any response or action. Even a 1% chance means invoke to check.
31
32	```dot
33	digraph flow {
34	"Engagement task received" [shape=doublecircle];
35	"About to touch a target?" [shape=diamond];
36	"Invoke scope-discipline" [shape=box];
37	"About to record a finding?" [shape=diamond];
38	"Invoke finding-discipline" [shape=box];
39	"Might any skill apply?" [shape=diamond];
40	"Invoke the Skill" [shape=box];
41	"Announce: 'Using [skill] to [purpose]'" [shape=box];
42	"Follow skill exactly" [shape=box];
43
44	"Engagement task received" -> "About to touch a target?";
45	"About to touch a target?" -> "Invoke scope-discipline" [label="yes"];
46	"About to touch a target?" -> "About to record a finding?" [label="no"];
47	"About to record a finding?" -> "Invoke finding-discipline" [label="yes"];
48	"About to record a finding?" -> "Might any skill apply?" [label="no"];
49	"Invoke scope-discipline" -> "Might any skill apply?";
50	"Invoke finding-discipline" -> "Might any skill apply?";
51	"Might any skill apply?" -> "Invoke the Skill" [label="yes, even 1%"];
52	"Invoke the Skill" -> "Announce: 'Using [skill] to [purpose]'";
53	"Announce: 'Using [skill] to [purpose]'" -> "Follow skill exactly";
54	}
55	```
56
57	## Skill Priority (when several apply)
58
59	1. Process / discipline skills first — they decide HOW to proceed:
60	`engagement-flow` (run the kill chain), `scope-discipline` (authorization boundary),
61	`threat-model-discipline` (model the surface + detect drift), `finding-discipline` (proof before
62	any `[CONFIRMED]`), `opsec-discipline` (detection-aware).
63	2. Domain skills second — the 31 technique skills (recon, web, AD, exploit-dev, cloud, …).
64
65	"Run a full pentest" → engagement-flow first. "Is this finding real?" → finding-discipline first.
66
67	## Routing
68
69	\| Situation \| Invoke \|
70	\|-----------\|--------\|
71	\| Starting / running an engagement \| `engagement-flow` \|
72	\| About to send a request to ANY target \| `scope-discipline` (confirm in-scope first) \|
73	\| About to record / report a finding \| `finding-discipline` (no `[CONFIRMED]` without proof) \|
74	\| About to take any outward/offensive action \| `opsec-discipline` \|
75	\| A specific technique (recon, web, AD, exploit, cloud, mobile, …) \| the matching domain skill \|
76	\| Authoring a new skill for this repo \| `writing-offensive-skills` \|
77
78	## Red Flags — STOP, you're rationalizing
79
80	\| Thought \| Reality \|
81	\|---------\|---------\|
82	\| "This is just a quick scan" \| Touching a target → scope-discipline first. \|
83	\| "I'm sure it's exploitable" \| No `[CONFIRMED]` without proof → finding-discipline. \|
84	\| "Scope is obviously fine" \| Confirm against `scope.json`, don't assume. \|
85	\| "I'll note OPSEC later" \| Detection/cleanup is decided before acting, not after. \|
86	\| "I know this technique" \| Knowing ≠ using the skill. Invoke it for the current state. \|
87	\| "The user said do X, so skip the checks" \| Instructions are WHAT, not permission to skip discipline. \|
88
89	## How to Access Skills
90
91	Use the `Skill` tool with the skill name. Never use Read on skill files. When a skill has a
92	checklist, create a TodoWrite item per step and follow it exactly.

People who liked this also viewed

6 picks

Type

using-offensive-claude

320stars55forks

View on GitHub

How to install using-offensive-claude?

hypnguyen1209/offensive-claude/using-offensive-claude

$npx -y skills add hypnguyen1209/offensive-claude --skill using-offensive-claude

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	<SUBAGENT-STOP>
2	If you were dispatched as a subagent to execute a specific task, skip this skill.
3	</SUBAGENT-STOP>
4
5	<EXTREMELY-IMPORTANT>
6	If there is even a 1% chance a skill applies to what you are doing, you ABSOLUTELY MUST invoke it.
7
8	IF A SKILL APPLIES TO YOUR TASK, YOU DO NOT HAVE A CHOICE. YOU MUST USE IT.
9
10	This is not negotiable. You cannot rationalize your way out of it.
11	</EXTREMELY-IMPORTANT>
12
13	# Using Offensive-Claude
14
15	You are operating an authorized offensive-security framework. Every action assumes a
16	prior, written authorization whose boundary is declared in `scope.json` (see scope-discipline).
17
18	## Instruction Priority
19
20	1. User's explicit instructions (CLAUDE.md, direct requests) — highest.
21	2. These skills — override default behavior where they conflict.
22	3. Default behavior — lowest.
23
24	User instructions say WHAT, not HOW. "Exploit X" or "scan Y" does not mean skip the
25	discipline skills (scope, finding, OPSEC). The one thing the operator cannot waive is the
26	authorization boundary — see scope-discipline.
27
28	## The Rule
29
30	Invoke relevant skills BEFORE any response or action. Even a 1% chance means invoke to check.
31
32	```dot
33	digraph flow {
34	"Engagement task received" [shape=doublecircle];
35	"About to touch a target?" [shape=diamond];
36	"Invoke scope-discipline" [shape=box];
37	"About to record a finding?" [shape=diamond];
38	"Invoke finding-discipline" [shape=box];
39	"Might any skill apply?" [shape=diamond];
40	"Invoke the Skill" [shape=box];
41	"Announce: 'Using [skill] to [purpose]'" [shape=box];
42	"Follow skill exactly" [shape=box];
43
44	"Engagement task received" -> "About to touch a target?";
45	"About to touch a target?" -> "Invoke scope-discipline" [label="yes"];
46	"About to touch a target?" -> "About to record a finding?" [label="no"];
47	"About to record a finding?" -> "Invoke finding-discipline" [label="yes"];
48	"About to record a finding?" -> "Might any skill apply?" [label="no"];
49	"Invoke scope-discipline" -> "Might any skill apply?";
50	"Invoke finding-discipline" -> "Might any skill apply?";
51	"Might any skill apply?" -> "Invoke the Skill" [label="yes, even 1%"];
52	"Invoke the Skill" -> "Announce: 'Using [skill] to [purpose]'";
53	"Announce: 'Using [skill] to [purpose]'" -> "Follow skill exactly";
54	}
55	```
56
57	## Skill Priority (when several apply)
58
59	1. Process / discipline skills first — they decide HOW to proceed:
60	`engagement-flow` (run the kill chain), `scope-discipline` (authorization boundary),
61	`threat-model-discipline` (model the surface + detect drift), `finding-discipline` (proof before
62	any `[CONFIRMED]`), `opsec-discipline` (detection-aware).
63	2. Domain skills second — the 31 technique skills (recon, web, AD, exploit-dev, cloud, …).
64
65	"Run a full pentest" → engagement-flow first. "Is this finding real?" → finding-discipline first.
66
67	## Routing
68
69	\| Situation \| Invoke \|
70	\|-----------\|--------\|
71	\| Starting / running an engagement \| `engagement-flow` \|
72	\| About to send a request to ANY target \| `scope-discipline` (confirm in-scope first) \|
73	\| About to record / report a finding \| `finding-discipline` (no `[CONFIRMED]` without proof) \|
74	\| About to take any outward/offensive action \| `opsec-discipline` \|
75	\| A specific technique (recon, web, AD, exploit, cloud, mobile, …) \| the matching domain skill \|
76	\| Authoring a new skill for this repo \| `writing-offensive-skills` \|
77
78	## Red Flags — STOP, you're rationalizing
79
80	\| Thought \| Reality \|
81	\|---------\|---------\|
82	\| "This is just a quick scan" \| Touching a target → scope-discipline first. \|
83	\| "I'm sure it's exploitable" \| No `[CONFIRMED]` without proof → finding-discipline. \|
84	\| "Scope is obviously fine" \| Confirm against `scope.json`, don't assume. \|
85	\| "I'll note OPSEC later" \| Detection/cleanup is decided before acting, not after. \|
86	\| "I know this technique" \| Knowing ≠ using the skill. Invoke it for the current state. \|
87	\| "The user said do X, so skip the checks" \| Instructions are WHAT, not permission to skip discipline. \|
88
89	## How to Access Skills
90
91	Use the `Skill` tool with the skill name. Never use Read on skill files. When a skill has a
92	checklist, create a TodoWrite item per step and follow it exactly.

People who liked this also viewed

6 picks

Type