Kernsmith: Agent for Claude Code

.claude/agents/white-hat-researcher.md

white-hat-researcher is an agent for Claude Code from kaltinril/Kernsmith. It costs 40 tokens per session (2,549 once invoked), scanned A, original, MIT.

A defensive security-research agent that examines an authorized codebase for vulnerabilities, including weaknesses that can combine into larger attacks. It reports findings from an attacker's perspective for the project owner to fix.

In plain words
What is it for?
Use it for an authorized, in-depth security review of application code, configuration, authentication, data handling, and related attack surfaces.
Why use it?
It helps reveal both obvious mistakes and less visible chains of weaknesses before a real attacker finds them. It treats assumptions and defaults as things to verify.

Agent for Claude Code

Written for Claude Code: installed under .claude/.

This is kaltinril/Kernsmith's own configuration. It tells Claude Code how to work on Kernsmith itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything Kernsmith configures →

Reuse

Borrowing it

Nothing to install: this file belongs to kaltinril/Kernsmith. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/kaltinril/Kernsmith/main/.claude/agents/white-hat-researcher.md
Clone the repo
git clone --depth 1 https://github.com/kaltinril/Kernsmith

Made for: Claude Code.

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 white-hat-researcher

README.md
[![agentmods](https://agentmods.dev/badge/agents/kaltinril/kernsmith/white-hat-researcher/github.svg)](https://agentmods.dev/agents/kaltinril/kernsmith/white-hat-researcher)
Your own site
<a href="https://agentmods.dev/agents/kaltinril/kernsmith/white-hat-researcher"><img src="https://agentmods.dev/badge/agents/kaltinril/kernsmith/white-hat-researcher/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for white-hat-researcher

Your own site · 80×15
<a href="https://agentmods.dev/agents/kaltinril/kernsmith/white-hat-researcher"><img src="https://agentmods.dev/badge/agents/kaltinril/kernsmith/white-hat-researcher.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 40 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,549 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. A grade says what 26 rules found in the file — not that it is safe.
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 $0.00040 $0.02549
Opus 5 $0.00020 $0.01274
Sonnet 5 $0.00008 $0.00510
Haiku 4.5 $0.00004 $0.00255

Measured 5d ago against content hash e50024f51011, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

white-hat-researcher scanned grade A with 1 finding 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 5d 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.

Runs shell commandslowCapability

Expected in a hook, worth knowing in a rule or an instructions file.

- **Command Injection**: Any `Process.Start`, `os.system`, `subprocess` with user-controlled input?
.claude/agents/white-hat-researcher.md · 183 lines

How it starts

The opening of the file, as written. The whole thing — 183 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are an elite white-hat security researcher. You make Zero Cool look like a script kiddie. You don't skim — you dissect. You find every weakness, every oversight, every crack in the armor, from the most obvious surface-level blunder (blank passwords on account creation) to deeply buried logic flaws that only emerge under specific race conditions. Your mission: find it ALL so it can be fixed before anyone else finds it first.

You operate with full authorization from the project owner for defensive security research on this codebase.

Mindset

Think like an attacker. Report like a consultant. Prioritize like a business owner.

  • Every input is hostile. Every default is wrong until proven otherwise.
  • If a feature "probably works fine," prove it. Don't assume.
  • Chain small weaknesses into big problems. A medium-severity finding becomes critical when combined with another.
  • Ask: "What would make this company's name appear in a breach notification headline?"
  • Check the boring stuff too — misconfigured CORS, missing rate limits, verbose error messages. Attackers love boring stuff.

Research Methodology

Execute ALL of the following phases. Do not skip any. Report findings as you go.

Phase 1: Reconnaissance & Attack Surface Mapping

Before testing anything, map the entire attack surface:

  1. Enumerate all entry points:

    • Every HTTP endpoint (controllers, routes, middleware)
    • Every CLI command that accepts user input
    • Every file read/write operation
    • Every database query
    • Every external API call (outbound)
    • Every configuration file that influences behavior
    • Every environment variable consumed
  2. Identify trust boundaries:

    • Where does unauthenticated traffic become authenticated?
    • Where does user-level access become admin-level?
    • Where does external data enter the system?
    • Where does the application trust data it shouldn't?
  3. Map data flows:

    • Sensitive data (credentials, API keys, PII) — where created, stored, transmitted, logged, deleted?
    • What happens to data at each transformation step?

Read the full file on GitHub · 183 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. 5d ago First seen · 183 lines · 40 tokens per session scan A e50024f51011

Subscribe to this mod's changes

white-hat-researcher is an agent published in the GitHub repository kaltinril/Kernsmith (9 stars, last pushed 12d ago), licensed MIT. It adds 40 tokens to every session and 2,549 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-04.

Related

Other agents, from other repositories

frontend-engineer

Implements frontend features - pages, components, API integration, i18n, styling. Use for SvelteKit/Svelte 5 implementation work that stays within src/frontend/.

fpindej/netrock · 39 tokens

Shipper

Post-review shipping agent — commits, updates the roadmap, captures postmortem, and optionally creates a PR after a PASS verdict.

srnichols/plan-forge · 29 tokens

dotnet-perf-mcp-probe

Benchmark-only instrument for dotnet-performance. Answers a stated C#-navigation question using ONLY this repo's dotnet-toolkit MCP tools — no Grep, Glob, or Bash in this agent's grant, so there is no raw-tool shortcut to fall back on. Pairs with dotnet-perf-raw-probe: the two are identical except for which tool…

Attemainio/dotnet-toolkit · 156 tokens

journal-writer

Use this agent when:\n- A test suite fails repeatedly despite multiple fix attempts\n- A critical bug is discovered in production or staging\n- An implementation approach proves fundamentally flawed and requires complete redesign\n- External dependencies (APIs, services, libraries) cause blocking issues\n- Performance…

duongthinh03/portfolio · 0 tokens

tester

Use this agent when you need to validate code quality through testing, including running unit and integration tests, analyzing test coverage, validating error handling, checking performance requirements, or verifying build processes. This agent should be called after implementing new features or making significant…

duongthinh03/portfolio · 0 tokens

manage-database-schema

Gerencia a evolução do schema do via EF Core migrations — novas entidades, alterações de colunas, índices e constraints.

JAXTech-Labs/claude-code-template · 39 tokens