Safety notice: Source 1f56ad9 (2026-09-23): run_script() uses subprocess shell=True with no allowlist, the bots accept tasks from any sender, and the advertised tools are never called by the agent loop, so the listing's Nmap/ClamAV/automation features do not function as sold. A security vendor must not merchandise this; buyers see the finding. This product is not recommended or promoted by the store until the finding is fixed.
CYBER-CLAW Agent Policy Kit — Agent loop template with human approval — $49 · ₹4,067 Software · AI Agent & MCP Security · AI Agent & LLM Security

CYBER-CLAW Agent Policy Kit

UnverifiedSoftware

Agent loop template with human approval — $49 · ₹4,067

Best for: Builders wanting a local agent-loop template (LangGraph + Ollama) with human-in-the-loop control.

A local AI agent that runs 24/7 on your own machine — agentic task-chaining via LangGraph, offline LLM support through Ollama, and Telegram/Discord control — paired with a built-in security engine (Nmap, ClamAV, AI anomaly detection) for automation and security ops in one process.

What it does

  • Agentic AI via LangGraph — task chaining and multi-agent workflows
  • Local/offline LLM support (Ollama) — no cloud dependency required
  • Telegram and Discord control for remote operation
  • Task automation across browser, email, social, files, and scripts
  • Built-in security engine — Nmap, ClamAV, AI anomaly detection
  • Encrypted config and a secure sandbox for execution

Self-serve at list. Seller: CYBERDUDEBIVASH ECOSYSTEM · GSTIN 21ARKPN8270G1ZP · PAN ARKPN8270G.

Gumroad receipt is not a GST tax invoice. ISO 27001 / SOC 2: aligned, not certified. No fabricated uptime, customer-count, or RPS guarantee.

Before you buy

Software facts

Every line comes from the product's evidence record. Where the store has no evidence, the row is left out rather than guessed. Unverified

What this isSoftware
Best forBuilders wanting a local agent-loop template (LangGraph + Ollama) with human-in-the-loop control.
Deployment modelSelf-hosted CLI (runs on your machine)
Platform / requirementsPython (single script) + Ollama
Store inspectioncompiles / imports cleanly; 1 high-severity static findings
LicenseProprietary, all rights reserved. (ships in the download)
Updateslifetime updates
SupportEmail contact@cyberdudebivash.in (no guaranteed response time)
Refund30-day refund (per Gumroad listing summary).

Known limitations

  • In the v2.5 source the agent never calls its tools: agent_node() returns the model's text only, so the Nmap, ClamAV, email, social-posting, file and script functions are not reachable from chat.
  • run_script() executes its input with subprocess shell=True and no command allowlist. Wiring the tools in without an allowlist and a sender allowlist would give anyone who can message the bot a shell on your machine.
  • The Telegram and Discord handlers accept tasks from any sender (no user allowlist).
  • No automated test suite ships with this product yet.
  • 1 high-severity static-analysis finding(s) in the source — see the evidence notes before use.
  • The store has not inspected the delivered files for this SKU; the description is the seller's listing with unsupported claims removed.
FAQ

Common questions

What exactly do I get when I buy CYBER-CLAW Agent Policy Kit?

The software files as delivered by Gumroad on payment. License: Proprietary, all rights reserved.

Is this a subscription?

No. It is a one-time purchase with no recurring charge. Updates: lifetime updates.

Has the store verified this product?

Unverified: The store has not inspected the delivered files. Description is the seller's listing, edited to remove unsupported claims.

What is the refund policy?

This product's listing states: 30-day refund (per Gumroad listing summary). That term applies to this product instead of the store's default 7-day policy. See the Refund Policy.

How do I get support or ask a question before buying?

Email contact@cyberdudebivash.in with "CYBER-CLAW Agent Policy Kit" in the subject. For an existing order include your Gumroad order number. No guaranteed response time is offered for self-serve purchases.

AI Agent & MCP Security
Source inspectedSoftware

AGENTGUARD Policy Kit

Policy kit for tool-calling agents — $79 · ₹6,557

Best for: AI engineers adding prompt-injection, tool-allowlist and exfil guardrails to a tool-calling agent.

$79 · ₹6,557 one-time View →
AI Agent & MCP Security
Assessment kitAssessment kit

Prompt Injection Defense Kit

Control catalog and tests for LLM apps — ₹1,999 / book $24

Best for: AppSec/AI teams wanting a prompt-injection attack catalog, detection rules and IR playbook to adopt.

₹1,999 · $24 one-time View →
AI Agent & MCP Security
Assessment kitAssessment kit

AI Security Red Team Toolkit

Red-team worksheet pack — ₹2,999 / book $36

Best for: Consultants running authorized AI red-team assessments who want a worksheet/methodology pack.

₹2,999 · $36 one-time View →