Skip to content

Commit d03eb08

Browse files
authored
Merge pull request #5 from SnailSploit/claude/add-features-professionalism-gfVki
Claude/add features professionalism gf vki
2 parents 40ce4cf + c7b211b commit d03eb08

67 files changed

Lines changed: 6676 additions & 83 deletions

File tree

Some content is hidden

Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.

CHANGELOG.md

Lines changed: 49 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,49 @@
1+
# Changelog
2+
3+
All notable changes to `claude-red` are documented here. The library follows a phased roadmap (see [README.md](README.md#roadmap)). Versions follow [Semantic Versioning](https://semver.org/) where breaking changes mean skill renames, removals, or category restructures.
4+
5+
## [Unreleased]
6+
7+
### Planned
8+
9+
- Phase 1 — Internal AD/Windows split (16 skills)
10+
- Phase 2 — Cloud Identity / Hybrid (10 skills)
11+
- Phase 3 — Wireless split (12 skills) — **mandatory for v0.3.0**
12+
- Phase 4 — IoT split (10 skills)
13+
- Phase 5 — Web Basics (8 skills)
14+
- Phase 6 — Web Advanced (10 skills)
15+
16+
## [0.2.0] — 2025-05
17+
18+
### Added
19+
20+
- 7 new offensive skills:
21+
- `offensive-active-directory` — AD attack methodology (Kerberoast, ASREProast, ACL abuse, ADCS ESC1-15, delegation, persistence, hybrid AAD)
22+
- `offensive-wifi` — 802.11 attacks (WPA2/WPA3, EAP, KARMA/Mana, KRACK, WPS, BLE, Zigbee, Z-Wave, LoRa, sub-GHz)
23+
- `offensive-business-logic` — workflow bypass, price/coupon/refund abuse, race conditions, anti-fraud defeat, chain construction
24+
- `offensive-toctou` — time-of-check/use across binary, kernel, web, container layers with window-widening primitives
25+
- `offensive-iot` — hardware recon, firmware extraction, RTOS, ICS/OT, wireless protocols, MQTT/CoAP
26+
- `offensive-mobile` — Android+iOS pentest (Frida, pinning bypass, storage, biometric, deep links, Firebase) [category-sized]
27+
- `offensive-cloud` — AWS+Azure+GCP attack paths (privesc, IMDS, cross-account, persistence, CSPM evasion) [category-sized]
28+
- `offensive-reporting` — pro pentest report writing methodology (CVSS scoring, evidence hygiene, executive summaries, finding templates, attack chain narratives, deliverable formats, retest discipline)
29+
30+
### Changed
31+
32+
- **Reorganized skills into 13 category subdirectories.** Top-level `Skills/` now contains category folders (`web/`, `auth/`, `active-directory/`, `wireless/`, `cloud/`, `mobile/`, `iot/`, `infrastructure/`, `exploit-dev/`, `fuzzing/`, `recon/`, `ai/`, `utility/`). Skill folder names unchanged.
33+
- README rewritten with category-based navigation, badges, install snippets, and roadmap.
34+
- SECURITY.md rewritten with intended-use scope and disclosure policy.
35+
36+
### Added (Documentation & Packaging)
37+
38+
- `LICENSE` — MIT (was claimed in README, file now present)
39+
- `CONTRIBUTING.md` — skill format, frontmatter standard, review process
40+
- `CHANGELOG.md` — this file
41+
- `claude-skills.json` — machine-readable manifest of all skills
42+
- `install.sh` — installer that copies skills into the Claude skills path
43+
44+
## [0.1.0] — 2024
45+
46+
### Added
47+
48+
- Initial library of 37 offensive security skills, derived from the SnailSploit / Sahar Shlichov offensive checklist collection
49+
- Categories covered: web app, auth, infrastructure & binary, recon, fuzzing, AI security, utility

CONTRIBUTING.md

Lines changed: 140 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,140 @@
1+
# Contributing to claude-red
2+
3+
Thanks for contributing. This guide explains the skill format, the review process, and the conventions to keep the library coherent as it grows.
4+
5+
## Quick Rules
6+
7+
1. **One skill, one surface.** Prefer focused skills (`offensive-kerberoasting`) over monolithic overviews (`offensive-active-directory`).
8+
2. **YAML frontmatter is required.** Skills without it won't load via the Claude Skills system.
9+
3. **Cite sources.** Every technique should be attributable. Link CVEs, advisories, original research.
10+
4. **No unauthorized targeting.** Don't include hardcoded victim domains, real customer data, or credentials.
11+
5. **Use code blocks with language tags.** It's how Claude (and humans) parse them best.
12+
13+
---
14+
15+
## Skill Format
16+
17+
A skill lives at:
18+
19+
```
20+
Skills/<category>/<skill-folder>/SKILL.md
21+
```
22+
23+
The folder name **must** match the `name:` field in the frontmatter.
24+
25+
### Frontmatter (required)
26+
27+
```yaml
28+
---
29+
name: offensive-<bug-class-or-domain>
30+
description: "One paragraph (50–500 words). State the surface, the techniques covered, and when to use this skill. Claude uses this for trigger matching — be specific about scenarios, tools, and sub-topics."
31+
---
32+
```
33+
34+
The `description` is what Claude matches against. Make it dense with relevant terms an operator would mention. Avoid marketing language.
35+
36+
### Body Structure (recommended)
37+
38+
```markdown
39+
# <Short Skill Title>
40+
41+
<One-paragraph framing. Why this matters, what makes it distinct.>
42+
43+
## Quick Workflow
44+
45+
1. <Numbered, ordered steps an operator follows in the field>
46+
47+
---
48+
49+
## <Section per phase or technique cluster>
50+
51+
<Concrete, copy-paste commands or code blocks. Annotate the why.>
52+
53+
---
54+
55+
## Detection / Defender View
56+
57+
<Optional but valuable — what defenders will see, common evasions.>
58+
59+
---
60+
61+
## Engagement Cheatsheet
62+
63+
<A short copy-paste-ready sequence summarizing the methodology.>
64+
65+
---
66+
67+
## Key References
68+
69+
- MITRE ATT&CK / CWE / OWASP IDs
70+
- Canonical research papers, conference talks
71+
- Tool docs, advisory URLs
72+
- Source: link to upstream checklist if applicable
73+
```
74+
75+
### Style Guide
76+
77+
- **Voice:** technical, second-person ("you"), present tense
78+
- **Length:** 200–800 lines is typical; aim for depth in one surface, not breadth across many
79+
- **Code blocks:** always specify the language (`bash`, `python`, `c`, `powershell`, `sql`, `yaml`, `http`)
80+
- **Tables:** use for compact reference (CVE → exploitation, capability → escape, etc.)
81+
- **No emoji** unless used as visual markers in tables (✓ ✗ ⚠) and only sparingly
82+
83+
---
84+
85+
## Adding a New Skill
86+
87+
1. Pick the right category folder. If none fits, propose a new one in your PR description.
88+
2. Create `Skills/<category>/<skill-name>/SKILL.md`.
89+
3. Write the frontmatter and body following the structure above.
90+
4. Update [`README.md`](README.md) — add the skill to the relevant category table.
91+
5. Update [`CHANGELOG.md`](CHANGELOG.md) under the next version.
92+
6. Update [`claude-skills.json`](claude-skills.json) if it exists (run `python tools/build_manifest.py` if available).
93+
7. Run any local lint:
94+
```bash
95+
./tools/check-skill.sh Skills/<category>/<skill-name>/SKILL.md
96+
```
97+
98+
## Modifying an Existing Skill
99+
100+
- Preserve the `name:` field (it's a public identifier; renames are breaking changes)
101+
- Note the edit briefly in CHANGELOG.md
102+
- For substantive rewrites, link the prior version's SHA so reviewers can diff
103+
104+
## Splitting a Monolithic Skill
105+
106+
When a skill grows beyond one surface (e.g. `offensive-wifi` covering WPA2, WPA3, BLE, Zigbee), split it:
107+
108+
1. Keep the original as a brief overview that points to the new focused skills
109+
2. Move detailed content into new per-surface skills
110+
3. Update README, CHANGELOG, and the manifest
111+
112+
The roadmap in README tracks current splits.
113+
114+
## Review Process
115+
116+
Pull requests are reviewed for:
117+
118+
- Technical accuracy (does this work? is it current?)
119+
- Clarity (would a competent operator understand and execute?)
120+
- Scope (one surface, not three)
121+
- Attribution (sources cited?)
122+
- Safety (no real targets, no live secrets, no malicious helpers)
123+
124+
Expect one round of review. Maintainers may request edits before merging.
125+
126+
---
127+
128+
## What We Won't Accept
129+
130+
- Skills that hardcode real victim infrastructure
131+
- Tooling that has destructive defaults without warnings
132+
- Bypasses for vendor-mandated security telemetry without legitimate red team context
133+
- Content under non-MIT-compatible licenses
134+
- AI-generated skills without operator review (use Claude to draft, then verify and edit)
135+
136+
---
137+
138+
## Questions
139+
140+
Open a GitHub Discussion before a large PR so the maintainers can confirm the direction. For sensitive findings (a leaked credential in an example, etc.), see [SECURITY.md](SECURITY.md).

LICENSE

Lines changed: 21 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,21 @@
1+
MIT License
2+
3+
Copyright (c) 2024-2025 SnailSploit / Kai Aizen
4+
5+
Permission is hereby granted, free of charge, to any person obtaining a copy
6+
of this software and associated documentation files (the "Software"), to deal
7+
in the Software without restriction, including without limitation the rights
8+
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
9+
copies of the Software, and to permit persons to whom the Software is
10+
furnished to do so, subject to the following conditions:
11+
12+
The above copyright notice and this permission notice shall be included in all
13+
copies or substantial portions of the Software.
14+
15+
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
16+
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
17+
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
18+
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
19+
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
20+
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
21+
SOFTWARE.

0 commit comments

Comments
 (0)