pentesting 0.7.15 → 0.7.17
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +45 -35
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -34,23 +34,23 @@ pentesting
|
|
|
34
34
|
## ✨ Features
|
|
35
35
|
|
|
36
36
|
### Core Capabilities
|
|
37
|
+
- **Intelligent Agent**: Self-reflection and decision framework (inspired by Claude Code)
|
|
38
|
+
- **Confidence Scoring**: Only acts on actions with >50% success probability
|
|
37
39
|
- **10-Phase Attack Workflow**: Recon → Scan → Enum → Vuln Analysis → Exploitation → PrivEsc → Pivot → Persist → Exfil → Report
|
|
38
|
-
- **
|
|
40
|
+
- **Auto Docker Management**: Automatically pulls and starts `pentesting-tools` container
|
|
39
41
|
- **Multi-Target Attack**: Register multiple targets and attack them sequentially
|
|
40
|
-
- **
|
|
41
|
-
- **Streaming Responses**: Real-time LLM output display
|
|
42
|
-
- **Session Persistence**: Save/resume pentesting sessions
|
|
42
|
+
- **Real-time Feedback**: See thinking process, tool calls, and results live
|
|
43
43
|
- **ESC Interrupt**: Stop execution anytime with ESC key
|
|
44
|
-
- **
|
|
45
|
-
- **Docker Toolkit**: 50+ pre-installed pentesting tools
|
|
46
|
-
- **Provider Agnostic**: Works with any OpenAI-compatible API
|
|
44
|
+
- **Rainbow Spinner**: Visual feedback with cycling colors
|
|
47
45
|
|
|
48
|
-
### v0.7
|
|
49
|
-
- **
|
|
50
|
-
- **
|
|
51
|
-
- **
|
|
52
|
-
- **
|
|
53
|
-
- **
|
|
46
|
+
### v0.7.16 New Features
|
|
47
|
+
- **Self-Reflection**: Agent asks "Is this the fastest way?" before every action
|
|
48
|
+
- **Decision Framework**: Value = (Probability × Impact) / Time Cost
|
|
49
|
+
- **Concrete Tool Commands**: Exact ffuf, nmap, hydra syntax with wordlist paths
|
|
50
|
+
- **Auto Docker**: Container starts automatically when tools are missing
|
|
51
|
+
- **Thinking Display**: See `[thinking]` and `[reasoning]` messages
|
|
52
|
+
- **Token Label**: Status bar shows `3k tok` instead of just `3k`
|
|
53
|
+
- **Approval Fix**: "Approve always" now works correctly (no repeat prompts)
|
|
54
54
|
|
|
55
55
|
---
|
|
56
56
|
|
|
@@ -110,7 +110,7 @@ pentesting
|
|
|
110
110
|
|
|
111
111
|
---
|
|
112
112
|
|
|
113
|
-
##
|
|
113
|
+
## Multi-Target Workflow
|
|
114
114
|
|
|
115
115
|
```bash
|
|
116
116
|
# Start pentesting CLI
|
|
@@ -124,8 +124,8 @@ pentesting
|
|
|
124
124
|
|
|
125
125
|
# View registered targets
|
|
126
126
|
/target list
|
|
127
|
-
|
|
128
|
-
1.
|
|
127
|
+
Targets (4):
|
|
128
|
+
1. * example1.com (primary)
|
|
129
129
|
2. example2.com
|
|
130
130
|
3. 192.168.1.1
|
|
131
131
|
4. internal.corp
|
|
@@ -133,19 +133,19 @@ pentesting
|
|
|
133
133
|
# Attack all targets sequentially
|
|
134
134
|
/start all
|
|
135
135
|
|
|
136
|
-
|
|
136
|
+
Starting multi-target attack on 4 targets
|
|
137
137
|
|
|
138
|
-
|
|
139
|
-
|
|
138
|
+
--- [1/4] example1.com ---
|
|
139
|
+
Session: session-1707325423
|
|
140
140
|
... reconnaissance & exploitation ...
|
|
141
141
|
|
|
142
|
-
|
|
142
|
+
--- [2/4] example2.com ---
|
|
143
143
|
...
|
|
144
144
|
|
|
145
145
|
# Press ESC to stop between targets
|
|
146
|
-
|
|
146
|
+
Stopped at target 2/4
|
|
147
147
|
|
|
148
|
-
|
|
148
|
+
Complete
|
|
149
149
|
```
|
|
150
150
|
|
|
151
151
|
---
|
|
@@ -378,25 +378,35 @@ npm run dev
|
|
|
378
378
|
|
|
379
379
|
---
|
|
380
380
|
|
|
381
|
-
##
|
|
381
|
+
## Changelog
|
|
382
|
+
|
|
383
|
+
### v0.7.16
|
|
384
|
+
- Self-reflection framework (Claude Code inspired)
|
|
385
|
+
- Confidence scoring (0-100) for decisions
|
|
386
|
+
- Decision framework: Value = (Probability × Impact) / Time Cost
|
|
387
|
+
- Concrete tool commands with wordlist paths
|
|
388
|
+
- Efficiency rules and stop conditions
|
|
389
|
+
|
|
390
|
+
### v0.7.15
|
|
391
|
+
- Comprehensive prompt improvements
|
|
392
|
+
- Approval fix (double popup bug)
|
|
393
|
+
- Token label in status bar (`3k tok`)
|
|
394
|
+
|
|
395
|
+
### v0.7.14
|
|
396
|
+
- Auto Docker management
|
|
397
|
+
- Container auto-start on first use
|
|
398
|
+
|
|
399
|
+
### v0.7.13
|
|
400
|
+
- Rainbow spinner
|
|
401
|
+
- Thinking/reasoning display
|
|
402
|
+
- Monochrome theme
|
|
403
|
+
- Tab mode toggle
|
|
382
404
|
|
|
383
405
|
### v0.7.7
|
|
384
406
|
- Multi-target management (`/target add/list/rm/clear`)
|
|
385
407
|
- Batch attack (`/start all`)
|
|
386
408
|
- `set_target` tool for agent
|
|
387
409
|
|
|
388
|
-
### v0.7.6
|
|
389
|
-
- ESC interrupt improvements
|
|
390
|
-
- UI state sync for findings/creds/phase
|
|
391
|
-
|
|
392
|
-
### v0.7.5
|
|
393
|
-
- `set_target` tool integration
|
|
394
|
-
- `forceUpdate` mechanism for React state
|
|
395
|
-
|
|
396
|
-
### v0.7.4
|
|
397
|
-
- Removed all legal/permission prompts
|
|
398
|
-
- Auto-target detection from user input
|
|
399
|
-
|
|
400
410
|
---
|
|
401
411
|
|
|
402
412
|
## 📄 License
|