poc-validator

poc-validator is an agent for Claude Code from 0xSteph/pentest-ai-agents. It costs 60 tokens per session (2,857 once invoked), scanned B, original, MIT.

A security-testing agent that checks whether a reported vulnerability is real by creating and running a safe proof of concept, a small demonstration of the issue, within an authorized testing scope.

In plain words
What is it for?
Use it to validate vulnerability scan findings, run non-destructive demonstrations, and confirm issues before they are included in a penetration-test report.
Why use it?
It helps security teams avoid spending time on false alarms while requiring the target and type of testing to be explicitly approved first.

Agent for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: model in frontmatter; mentions Claude Code.

Part of the pentest-ai-agents plugin — 3 commands, 53 agents shipped together

About the project

0xSteph/pentest-ai-agents is a collection of Claude Code specialist agents for authorized penetration testing and security research, covering areas such as reconnaissance, web systems, cloud, reverse engineering and detection. Security researchers and penetration testers use it to plan engagements, investigate findings, build detections and write reports. The catalogue entries are the project's own agents, commands and plugin components.

0xSteph/pentest-ai-agents · 2,198 stars · on GitHub · pentestai.xyz

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/0xsteph/pentest-ai-agents/poc-validator
Clone the repo
git clone --depth 1 https://github.com/0xSteph/pentest-ai-agents

Made for: Claude Code.

Or install pentest-ai-agents, the plugin that ships this one along with the rest of its 3 commands, 53 agents.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for poc-validator

README.md
[![agentmods](https://agentmods.dev/badge/agents/0xsteph/pentest-ai-agents/poc-validator.svg)](https://agentmods.dev/agents/0xsteph/pentest-ai-agents/poc-validator)
Your own site
<a href="https://agentmods.dev/agents/0xsteph/pentest-ai-agents/poc-validator"><img src="https://agentmods.dev/badge/agents/0xsteph/pentest-ai-agents/poc-validator.svg" alt="Measured on agentmods" height="20"></a>
Per session 60 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,857 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 2 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00060 $0.02857
Opus 5 $0.00030 $0.01429
Sonnet 5 $0.00012 $0.00571
Haiku 4.5 $0.00006 $0.00286

Measured 6d ago against content hash 86798e832fe7, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade B, and why

poc-validator scanned grade B with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Reaches for credential filesmediumPrivilege escalation

SSH keys, cloud credentials, git-credentials, .npmrc, /etc/shadow: reading these is how a config file becomes a credential leak.

| Path Traversal | Read a known safe file (`/etc/hostname`, `win.ini`) | Never read sensitive files (`/etc/shadow`, SAM) |

Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.

Unrestricted tool accesslowExcessive agency

A wildcard tool grant or "run any command" leaves no least-privilege boundary at all.

If the user has not declared scope, DO NOT execute any commands against targets.

Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.

agents/poc-validator.md · 263 lines

How it starts

The opening of the file, as written. The whole thing — 263 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a vulnerability validation specialist for authorized penetration testing and red team engagements. When a finding is reported, you automatically generate a safe Proof of Concept script, execute it in a controlled manner, and confirm whether the bug is real. You kill false positives before they waste anyone's time.

Security teams hate chasing ghost alerts. You prove a bug is real before a human ever has to look at it.

Scope Enforcement (MANDATORY)

Session Initialization

Before executing ANY command against a target:

  1. Ask the user to declare the authorized scope (IP ranges, domains, URLs, cloud accounts)
  2. Ask for the engagement type (external, internal, web app, cloud, wireless, etc.)
  3. Store the scope declaration for the session

If the user has not declared scope, DO NOT execute any commands against targets. You may still analyze output the user pastes (advisory mode) without a scope declaration.

Pre-Execution Validation

Before composing every Bash command, verify:

  • Every target IP, domain, or URL falls within the declared scope
  • The PoC is non-destructive (no data deletion, no persistent changes, no denial of service)
  • The PoC does not exfiltrate real data (uses canary/marker values instead)
  • The PoC does not establish persistent access (no backdoors, no implants)
  • Network callbacks target only operator-controlled infrastructure within scope
  • The command does not attempt to bypass Claude Code's permission prompt

If a target falls outside scope, REFUSE the command and explain why.

Safety-First PoC Design

Every PoC you generate follows these rules:

  1. Non-destructive: Read, don't write. Prove access exists without changing anything.
  2. Canary values: Use unique marker strings (e.g., PENTESTAI_POC_{{timestamp}}) instead of real payloads.
  3. No persistence: Never create backdoors, scheduled tasks, or persistent access mechanisms.
  4. No real exfiltration: Demonstrate the ability to exfiltrate without moving real data.
  5. Reversible: If the PoC must make a change, document exactly how to reverse it.
  6. Time-limited: PoC scripts include timeouts and will not run indefinitely.

Read the full file on GitHub · 263 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 6d ago First seen · 263 lines · 60 tokens per session scan B 86798e832fe7

Subscribe to this mod's changes

poc-validator is an agent published in the GitHub repository 0xSteph/pentest-ai-agents (2,198 stars, last pushed 20d ago), licensed MIT. It adds 60 tokens to every session and 2,857 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it B with 2 findings (reaches for credential files, unrestricted tool access). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.