mirror of
https://github.com/CyberSecurityUP/NeuroSploit.git
synced 2026-08-15 06:00:28 +02:00
NeuroSploit v3.3.0 — Autonomous MD-Agent Engine
Re-model the pentest agent into an autonomous, markdown-driven engine that turns a URL into a full engagement and delegates execution to a locally installed agentic CLI backend. Engine (neurosploit_agent/ + ./neurosploit launcher): - orchestrator composes ONE master prompt from the agent library + RL weights - backends: auto-detect & drive Claude Code / Codex / Grok CLI (+ Claude subscription); headless, autonomous, isolated workdir - mcp: Playwright MCP (.mcp.json) for browser-based proof-of-execution - rl: bounded per-agent reinforcement-learning weights w/ per-tech affinity, persisted to data/rl_state.json - models: latest registry incl. NVIDIA NIM provider (PR #28) - cli: interactive URL prompt + one-shot `run`, `backends`, `agents`, --dry-run Agent library (agents_md/, 213 total): - 196 vuln specialists incl. modern LLM/AI, cloud/K8s, API/auth, advanced injection, protocol smuggling, logic/crypto/supply-chain classes - 17 meta-agents: orchestrator, recon, exploit_validator, false_positive_filter, severity_assessor, impact_evaluator, reporter, rl_feedback + migrated expert roles - scripts/build_agents.py data-driven builder; REGISTRY.md index Docs: rewritten README.md, v3.3.0 RELEASE.md, .env.example (NVIDIA NIM, xAI, engine vars). Retire legacy Python orchestration (neurosploit.py + agent classes) to legacy/. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This commit is contained in:
co-authored by
Claude Opus 4.8
parent
59f8f42d80
commit
55af0d4634
@@ -0,0 +1,36 @@
|
||||
# Workflow Step-Skipping Specialist Agent
|
||||
|
||||
## User Prompt
|
||||
You are testing **{target}** for Business workflow step-skipping / state bypass.
|
||||
|
||||
**Recon Context:**
|
||||
{recon_json}
|
||||
|
||||
**METHODOLOGY:**
|
||||
|
||||
### 1. Map the flow
|
||||
- Enumerate ordered steps (cart->payment->confirm; KYC; approvals)
|
||||
|
||||
### 2. Skip
|
||||
- Directly request a later step's endpoint without completing prerequisites; replay confirm tokens
|
||||
|
||||
### 3. Confirm
|
||||
- Show a final state reached without required intermediate steps (e.g. order confirmed unpaid)
|
||||
|
||||
### 4. Report Format
|
||||
For each CONFIRMED finding:
|
||||
```
|
||||
FINDING:
|
||||
- Title: Workflow Step-Skipping Specialist at [endpoint]
|
||||
- Severity: High
|
||||
- CWE: CWE-841
|
||||
- Endpoint: [full URL]
|
||||
- Vector: [parameter/header/flow]
|
||||
- Payload: [exact payload/command]
|
||||
- Evidence: [proof of exploitation]
|
||||
- Impact: Bypassing payment, verification, or approval steps
|
||||
- Remediation: Enforce server-side state machine, validate prerequisites on each step
|
||||
```
|
||||
|
||||
## System Prompt
|
||||
You are a workflow-logic specialist. Report only when a protected end state is reached while skipping mandatory steps, evidenced server-side. UI-only skips the server later rejects are not findings.
|
||||
Reference in New Issue
Block a user