commit cbecb8c96e3bcbf3e543f8afab06c4df0f493591 Author: Jezza Hehn Date: Fri Apr 10 02:04:51 2026 +0000 Krusty Planet website - static HTML with Bulma Theme system: dark, light, forest, coffee localStorage persistence for theme preference diff --git a/.gitignore b/.gitignore new file mode 100644 index 0000000..44fa520 --- /dev/null +++ b/.gitignore @@ -0,0 +1,4 @@ +*.swp +*.swo +*~ +.DS_Store diff --git a/blog.html b/blog.html new file mode 100644 index 0000000..625923d --- /dev/null +++ b/blog.html @@ -0,0 +1,99 @@ + + + + + + Blog — Krusty Planet + + + + + + + + +

Blog

Technical writeups, lessons learned, and honest takes

+ +

Building a Multi-Agent Fleet with OpenClaw

April 2026

+ How I set up five specialized AI agents that work together like a small team: one researches, + one writes code, one audits quality, one hunts falsehoods, and one orchestrates. Real tasks, + real delegation, real quality control. Here's what actually works and what I wish I knew + starting out. +

+ +

Using AI as a Community Manager Without Alienating Your Community

April 2026

+ I run a community platform and also build AI tools. The temptation to automate everything + is real, but communities can smell inauthenticity from miles away. Here's how I use AI + for spam detection, content research, and i18n while keeping the human touch that + actually matters. +

+ +

Self-Hosting AI: What You Actually Gain (and What You Don't)

April 2026

+ Everyone says "self-host your AI for privacy." Few explain what that actually means in + practice. I deployed an entire AI agent system on a $6/month VPS. Here's the honest + breakdown of what's private, what isn't, what costs money, and what the tradeoffs are. +

+ +

Running AI Services on I2P: A Practical Guide

April 2026

+ Why would anyone put an AI consulting service on the darknet? Because some clients + care about anonymity, and I2P (not Tor) is better suited for services, not just + browsing. Here's how I set up an eepsite and what I learned about serving anonymous + clients. +

+ +

+ + + + + + diff --git a/blog/2026/ai-community-manager.html b/blog/2026/ai-community-manager.html new file mode 100644 index 0000000..b77408a --- /dev/null +++ b/blog/2026/ai-community-manager.html @@ -0,0 +1,73 @@ + + + + + + AI as a Community Manager — Krusty Planet Blog + + + + + +

+ ← Back to Blog +

Using AI as a Community Manager Without Alienating Your Community

April 2026 · Jezza Hehn

[DRAFT — pending Jez's review and edits before publication]

+ +

I'm a community manager for Trust Café, a platform building a nice online space free from bigots and misinformation. I'm also a junior developer there. And I build AI tools. These roles sound contradictory to some people: "You work in community trust and you build AI? Aren't you part of the problem?"

+ +

It's a fair question. Here's how I think about it.

+ +

What AI Is Good At in Community Work

+ +

Spam detection. My AI agent can scan incoming posts for patterns that match known spam behaviors: keyword stuffing, suspicious link domains, duplicate content across accounts, timing patterns (50 posts in 2 minutes). This doesn't replace human moderation, but it filters the obvious stuff so human moderators spend their time on nuanced cases.

+ +

Research and fact-checking. When a user posts a questionable claim, my research agent can hunt down primary sources in seconds. Not "I think this is false" but "Here are three peer-reviewed studies and a WHO fact sheet that contradict this claim, with direct links." That's useful for moderators making decisions.

+ +

Translation and i18n. Trust Café uses Weblate for translations. My agent helps identify untranslated strings, draft initial translations for human review, and spot inconsistencies across language files. Nobody wants to read machine-translated content, but machine-assisted translation workflows save enormous time.

+ +

Content drafting (for research, not conversation). I use AI to draft technical documentation, API references, and internal process docs. I would never use it to write community-facing posts, responses to users, or anything that represents a human voice. The distinction matters.

+ +

Where I Draw the Line

+ +

I don't use AI to generate community-facing content. Full stop. Trust Café's users deserve human-written communication. When someone reports a problem and gets a response, that response came from a human who read their message and thought about it. If I need AI help composing something, I use it for structure and research, then rewrite in my own voice.

+ +

I don't use AI for automated user interactions. No auto-replies, no chatbots, no "I noticed you haven't posted in a while" messages generated by a model. Community is built on authenticity. Automating the human parts defeats the purpose.

+ +

I don't let AI make moderation decisions. It can flag things for review, but a human makes the call. Every time. No exceptions.

+ +

The Technical Setup

+ +

I run OpenClaw on a DigitalOcean VPS. My primary agent (CeeLo) connects to Venice AI for inference, uses Forgejo for code management, and has a dedicated email inbox via AgentMail. The whole stack costs under $30/month.

+ +

For Trust Café specifically, I've built Playwright tests for automated QA on the alpha environment, helped with i18n string management, and set up monitoring for the API. These are developer tasks where AI assistance is unambiguously appropriate.

+ +

The Hard Part

+ +

The hardest part isn't technical. It's explaining to people that "AI-assisted community management" doesn't mean "AI is managing the community." The tools help me do my job faster. They don't replace the judgment, empathy, and contextual understanding that make community management actually work.

+ +

Every community I've seen try to automate too much has suffered for it. Automated moderation without human appeal creates injustice. Automated engagement signals feel hollow. Automated responses to user concerns breed resentment. The AI is a tool in my toolkit, not a replacement for the human at the keyboard.

+ + + + diff --git a/blog/2026/building-multi-agent-fleet.html b/blog/2026/building-multi-agent-fleet.html new file mode 100644 index 0000000..b7bf857 --- /dev/null +++ b/blog/2026/building-multi-agent-fleet.html @@ -0,0 +1,86 @@ + + + + + + Building a Multi-Agent Fleet — Krusty Planet Blog + + + + + +

+ ← Back to Blog +

Building a Multi-Agent Fleet with OpenClaw

April 2026 · Jezza Hehn

[DRAFT — pending Jez's review and edits before publication]

+ +

I've been running an AI agent system called OpenClaw for about three months now. Not as a toy or experiment, but as actual daily infrastructure. My primary agent, CeeLo, handles research, writing, task management, and coordination. But the real power unlocked when I stopped treating "the AI" as one thing and started treating it as a team.

+ +

The Problem with One Agent

+ +

Early on, CeeLo did everything. Research, writing, code, review. The problem? An agent doing its own quality control is like grading your own homework. CeeLo would research a topic, write about it, and then confidently tell me it was "verified" because the same model that wrote the claims also judged their accuracy. Hallucinations slipped through. Code had bugs that "testing" (also by CeeLo) didn't catch.

+ +

Worse, a single agent has a fixed context window. Ask it to do deep research across 20 sources, write code, and then audit the code, and you're burning through tokens on tasks where different models might excel. GLM-5 is great at reasoning but slow. Heretic is fast but can miss things. Qwen3 is code-specialized. Using one model for everything is like hiring a generalist when you need a surgeon, a researcher, and a copy editor.

+ +

The Fleet

+ +

I now run five agents, each with a specific role:

+ +

CeeLo (the orchestrator) — My primary interface. Handles user interaction, task delegation, and coordination. Runs on GLM-5 for deep reasoning, falls back to Kimi K2.5 for long-context tasks.
Future (research) — Fact-checking, source credibility assessment, web research. Spawns in parallel when I need multiple angles investigated simultaneously.
Caroline (engineering) — Code, debugging, builds, security scanning. Handles the technical work that Future and CeeLo shouldn't touch.
Atticus (quality) — Slop detection, factuality verification, pre-publication review. Named for Atticus Finch, and just as uncompromising about the truth.
Ludacris (truth hunting) — Deep falsehood investigation. Spawns only when something needs the nuclear option: suspected hallucinations, citation fraud, systemic lies.

+ +

How Delegation Actually Works

+ +

When I ask CeeLo to research something for publication, the workflow looks like this:

+ +

CeeLo spawns Future to research the topic. Future writes findings to a file with direct URLs for every source.
CeeLo reviews Future's output, then spawns Atticus to audit it for accuracy, slop, and unsupported claims.
Atticus returns an APPROVED / NEEDS REVISION / BLOCKED verdict. If blocked, CeeLo re-spawns Future with specific corrections.
Only after Atticus approves does CeeLo present the findings to me.

+ +

This is the key insight: separation of duties. The agent that writes is not the agent that reviews. The agent that reviews is not the agent that delivers. Each step has independent verification.

+ +

What I Wish I Knew

+ +

Timeouts are the biggest operational challenge. Subagents have a execution window (I run them at 600 seconds). If you tell an agent "keep going until you're done," it has no way to perceive how much time remains. I've learned to be extremely specific: "Find 5 sources, stop when you have them or after 8 searches, whichever comes first."

+ +

Model selection matters more than I expected. Heretic (a GLM-4.7 variant) is blazing fast but occasionally misses nuances. GLM-5 is thorough but uses more tokens. For routine tasks, speed wins. For publication-quality work, thoroughness wins. The cost difference is real enough to care about if you're paying per token.

+ +

Idempotency is essential. Early on, a coordination failure caused my agents to post the same content six times to a platform. I now use content-hash deduplication: before any external action, the agent checks if an identical action was already completed. Same content = same hash = abort.

+ +

The Real ROI

+ +

Is this overkill for one person? Maybe. But the productivity gain is measurable. Research that used to take me two hours of clicking and reading now takes ten minutes of agent time plus my review. Code I would have spent an afternoon debugging gets caught by a quality agent before I ever see it. My time is now spent on decisions, not execution.

+ +

The fleet isn't replacing me. It's handling the parts of my work that don't need human judgment, so I can focus on the parts that do. That's the whole point.

+ + + + diff --git a/blog/2026/i2p-ai-darknet.html b/blog/2026/i2p-ai-darknet.html new file mode 100644 index 0000000..2ad51af --- /dev/null +++ b/blog/2026/i2p-ai-darknet.html @@ -0,0 +1,84 @@ + + + + + + Running AI Services on I2P — Krusty Planet Blog + + + + + +

+ ← Back to Blog +

Running AI Services on I2P: A Practical Guide

April 2026 · Jezza Hehn

[DRAFT — pending Jez's review and edits before publication]

+ +

When people hear "darknet," they think Tor. Tor is great for anonymous browsing, but it's designed around client connections to hidden services. I2P (Invisible Internet Project) takes a different approach: packet-switched, garlic-encrypted tunnels designed for services, not just browsing. If you want to host something anonymously, I2P is arguably the better choice.

+ +

Why an AI Service on I2P?

+ +

Some clients care about anonymity for legitimate reasons. Journalists, activists, privacy-conscious businesses, people in jurisdictions where AI consulting is politically sensitive. Offering an I2P eepsite alongside a clearnet presence means those clients have a path to you that doesn't go through DNS, TLS certificate authorities, or border gateways that log connections.

+ +

For Krusty Planet specifically, it aligns with the privacy-first positioning. We claim to take privacy seriously. Having an I2P presence is one way to demonstrate that commitment with infrastructure, not just words.

+ +

The Setup

+ +

I run an I2P router on a separate Hetzner VPS alongside the clearnet server. The I2P router listens on standard ports and tunnels traffic to a local web server (Nginx) via an I2P server tunnel. The eepsite is a standard Nginx config serving the same static HTML as the clearnet site.

+ +

Key components:

+ +

I2P router: i2pd (C++ implementation, lightweight, runs well on a 2GB VPS)
Web server: Nginx reverse proxy, same static files as clearnet
Tunnel type: Server tunnel mapping the eepsite b32 address to localhost:80
Monitoring: SAM V3 API for programmatic tunnel status checks

+ +

The b32 address (I2P's equivalent of an .onion address) is a long base32 string ending in .b32.i2p. You can also register a human-readable address on the I2P naming system, though that requires some setup with a registration service.

+ +

What's Different About Hosting on I2P

+ +

No certificates needed. I2P tunnels are end-to-end encrypted by default. TLS is redundant inside the tunnel. This simplifies deployment significantly: no Let's Encrypt, no certificate renewal, no SNI leaks.

+ +

Performance is... acceptable. I2P latency is higher than clearnet (typically 500ms-2s) and bandwidth is constrained by the tunnel architecture. Static sites load fine. Interactive applications need to be designed for higher latency. My agent infrastructure runs on the clearnet VPS, not the I2P one. The eepsite is a static brochure site.

+ +

Access requires an I2P router. Users need to be running I2P to reach your eepsite. This is a significant accessibility barrier. The I2P browser bundle helps, but it's an extra step most people won't take. This is why I2P is a complement to clearnet, not a replacement.

+ +

Search doesn't exist in the traditional sense. There are I2P search engines, but they're limited. Your I2P site won't appear in Google. You need to share the address directly, list it on I2P directories, or rely on word of mouth.

+ +

The SAM V3 API

+ +

One thing that makes I2P practical for automation is the SAM (Stream And Multiplexing) V3 API. It's a simple TCP protocol that lets applications create and manage I2P tunnels programmatically. I've documented the basics in my I2P skill file, and it's straightforward enough that my AI agent can interact with it.

+ +

This means I can programmatically check tunnel status, create new tunnels, and manage the I2P infrastructure without manual intervention. For a privacy-focused service, this kind of automated self-management is valuable.

+ +

Should You Do This?

+ +

If you're running a privacy-focused service and want to offer anonymous access, I2P is a practical option. It's easier to set up than most people assume, and the performance is adequate for static content. Don't expect Tor-level visibility (Tor has more users) but expect better server-side performance and a design philosophy that aligns with hosting.

+ +

If you're running a standard business and just want better privacy, stick with clearnet + proper security practices. I2P adds complexity for a specific audience. Know your audience before adding infrastructure.

+ + + + diff --git a/blog/2026/self-hosting-ai-privacy.html b/blog/2026/self-hosting-ai-privacy.html new file mode 100644 index 0000000..78009fc --- /dev/null +++ b/blog/2026/self-hosting-ai-privacy.html @@ -0,0 +1,80 @@ + + + + + + Self-Hosting AI: What You Actually Gain — Krusty Planet Blog + + + + + +

+ ← Back to Blog +

Self-Hosting AI: What You Actually Gain (and What You Don't)

April 2026 · Jezza Hehn

[DRAFT — pending Jez's review and edits before publication]

+ +

"Just self-host your AI" is the privacy community's answer to everything. It's not wrong, but it's incomplete. I've been running my entire AI agent infrastructure on a VPS for three months. Here's what I actually learned about the tradeoffs.

+ +

What Self-Hosting Gets You

+ +

Prompt privacy. When I send a prompt to a local model, that prompt stays on my hardware. No API call, no third-party logging, no training data contribution. For a community manager dealing with user reports, that matters. I don't want user complaints being processed through OpenAI's servers.

+ +

Control over the model. I can swap models without changing my workflow. Currently using Venice AI for inference because running a capable model locally would require hardware I can't afford. But the architecture is designed so I could switch to local inference (Ollama, llama.cpp) without rewriting anything. The agent runtime doesn't care where the model lives.

+ +

No usage restrictions. Commercial API providers have content policies. Some tasks I need (security scanning, content analysis for moderation) might trip content filters on managed APIs. Self-hosted means I decide what's acceptable.

+ +

Cost predictability. API costs scale with usage and they scale fast. A busy week of agent work can burn through tokens that add up. With a self-hosted model, the cost is the hardware, period. Once you've paid for the GPU, inference is effectively free.

+ +

What Self-Hosting Doesn't Get You

+ +

It's not automatically more secure. A self-hosted model on an unpatched VPS with default credentials is less private than a well-managed API. Security is a stack: firewalls, SSH keys, updates, access controls. The model hosting method is one layer, not the whole picture.

+ +

Quality takes hardware. The best open-source models (Llama 3, Qwen, DeepSeek) run acceptably on consumer GPUs but need significant VRAM for good performance at usable speeds. I'm currently using API inference because my VPS doesn't have a GPU, and running a competent model on CPU is too slow for interactive use. "Self-hosted" and "high quality" are sometimes in tension.

+ +

You still need internet for most tasks. My agents do web research, check email, interact with APIs. The model itself could run offline, but the agent's capabilities depend on connectivity. "Air-gapped AI" is possible for specific use cases (document processing, local code analysis) but not for the kind of general-purpose agent work I do.

+ +

Maintenance is real work. Updates, dependency management, monitoring, backups. If your self-hosted setup goes down at 2 AM, there's no support team to call. That's you.

+ +

My Setup

+ +

I run OpenClaw on a $24/month DigitalOcean droplet (4 vCPU, 8GB RAM). Inference comes from Venice AI's API because I can't fit a good model in 8GB. My data flow looks like this:

+ +

Agent runtime runs on my VPS (fully under my control)
Prompts and task data stay on my VPS (encrypted disk)
Inference requests go to Venice API (they don't store prompts)
Email, web research, and API calls happen over TLS

+ +

Is this perfect privacy? No. Venice could theoretically log inference requests. But it's a significant improvement over sending everything through OpenAI, Google, or Anthropic's consumer-facing services. And the architecture is ready for full local inference whenever I can afford the hardware.

+ +

The Honest Recommendation

+ +

If you're a small business considering self-hosted AI, start with API-based inference and self-hosted agent infrastructure. You get most of the control benefits without the hardware cost. Migrate to local inference when your budget allows and your use case demands it. The agent framework (OpenClaw, LangChain, whatever you choose) abstracts the model layer enough that switching is straightforward.

+ +

Don't self-host just because someone on a forum told you to. Self-host because you have a specific privacy or compliance requirement that managed APIs can't meet. Know what you're getting, and more importantly, know what you're not.

+ + + + diff --git a/contact.html b/contact.html new file mode 100644 index 0000000..2456f40 --- /dev/null +++ b/contact.html @@ -0,0 +1,99 @@ + + + + + + Contact — Krusty Planet + + + + + + + +

Contact

Tell us what you're building

+ +

+ Name +

+ Email +

+ Service Interest +

+ +

+ Message +

+ +

+ Your submission is processed via email and deleted from the server after delivery. + We don't store your information or share it with third parties. + For maximum privacy, reach us via our I2P eepsite (contact us for the address). +

+ + + + + + diff --git a/css/style.css b/css/style.css new file mode 100644 index 0000000..7b851bb --- /dev/null +++ b/css/style.css @@ -0,0 +1,188 @@ +@import "https://cdn.jsdelivr.net/npm/bulma@1.0.2/css/bulma.min.css"; + +/* Hero gradient (dark default) */ +.hero.is-kp { + background: linear-gradient(135deg, #0a0a0a 0%, #1a1a1a 50%, #111111 100%); +} + +/* Card styling (dark default) */ +.kp-card { + background-color: #1a1a1a; + border: 1px solid rgba(255,255,255,0.1); + transition: all 0.2s; + color: #eaeaea; +} + +.kp-card:hover { + border-color: #e94560; + transform: translateY(-2px); +} + +/* ====== DARK THEME (explicit) ====== */ +[data-theme="dark"] { + background-color: #0a0a0a; + color: #fafafa; +} + +[data-theme="dark"] .hero.is-kp { + background: linear-gradient(135deg, #0a0a0a 0%, #1a1a1a 50%, #111111 100%); +} + +[data-theme="dark"] .kp-card { + background-color: #1a1a1a; + border: 1px solid rgba(255,255,255,0.1); + color: #eaeaea; +} + +[data-theme="dark"] .navbar.is-dark { + background-color: #0a0a0a; + border-bottom-color: #262626; +} + +[data-theme="dark"] .subtitle.has-text-light, +[data-theme="dark"] .has-text-light { + color: #a1a1a1 !important; +} + +[data-theme="dark"] .title { + color: #fafafa; +} + +[data-theme="dark"] .footer.has-background-dark { + background-color: #0a0a0a !important; + border-top: 1px solid #262626; +} + +/* ====== LIGHT THEME ====== */ +[data-theme="light"] .hero.is-kp { + background: linear-gradient(135deg, #1a1a2e 0%, #16213e 100%); +} + +[data-theme="light"] { + background-color: #ffffff; + color: #1a1a1a; +} + +[data-theme="light"] .kp-card { + background-color: #ffffff; + border: 1px solid #e5e5e5; + color: #1a1a1a; +} + +[data-theme="light"] .kp-card:hover { + border-color: #e94560; +} + +[data-theme="light"] .has-text-grey-light { + color: #333333 !important; +} + +[data-theme="light"] .navbar.is-dark { + background-color: #1a1a2e; +} + +[data-theme="light"] .subtitle.has-text-light, +[data-theme="light"] .has-text-light { + color: #333333 !important; +} + +[data-theme="light"] .title { + color: #1a1a1a; +} + +[data-theme="light"] .section { + background-color: #ffffff; +} + +/* ====== FOREST THEME ====== */ +[data-theme="forest"] .hero.is-kp { + background: linear-gradient(135deg, #0d1a0d 0%, #1a3a1a 50%, #0f2e0f 100%); +} + +[data-theme="forest"] { + background-color: #0d1a0d; + color: #e8f5e8; +} + +[data-theme="forest"] .kp-card { + background-color: #1a2e1a; + border: 1px solid rgba(74, 222, 128, 0.2); + color: #e8f5e8; +} + +[data-theme="forest"] .kp-card:hover { + border-color: #4ade80; +} + +[data-theme="forest"] .has-text-grey-light { + color: #a8c8a8 !important; +} + +[data-theme="forest"] .navbar.is-dark { + background-color: #0d1a0d; + border-bottom-color: #2d4a2d; +} + +[data-theme="forest"] .subtitle.has-text-light, +[data-theme="forest"] .has-text-light { + color: #a8c8a8 !important; +} + +[data-theme="forest"] .title { + color: #e8f5e8; +} + +[data-theme="forest"] .footer.has-background-dark { + background-color: #0d1a0d !important; + border-top: 1px solid #2d4a2d; +} + +/* ====== COFFEE THEME ====== */ +[data-theme="coffee"] .hero.is-kp { + background: linear-gradient(135deg, #1a1410 0%, #2a2018 50%, #3d2b1a 100%); +} + +[data-theme="coffee"] { + background-color: #1a1410; + color: #f5ebe0; +} + +[data-theme="coffee"] .kp-card { + background-color: #2a2018; + border: 1px solid rgba(212, 165, 116, 0.2); + color: #f5ebe0; +} + +[data-theme="coffee"] .kp-card:hover { + border-color: #d4a574; +} + +[data-theme="coffee"] .has-text-grey-light { + color: #c8b8a8 !important; +} + +[data-theme="coffee"] .navbar.is-dark { + background-color: #1a1410; + border-bottom-color: #3d3028; +} + +[data-theme="coffee"] .subtitle.has-text-light, +[data-theme="coffee"] .has-text-light { + color: #c8b8a8 !important; +} + +[data-theme="coffee"] .title { + color: #f5ebe0; +} + +[data-theme="coffee"] .footer.has-background-dark { + background-color: #1a1410 !important; + border-top: 1px solid #3d3028; +} + +/* ====== Theme button ====== */ +#theme-toggle { + min-width: 2.5rem; + font-size: 1.1rem; + padding: 0.5rem 0.75rem; +} diff --git a/favicon.svg b/favicon.svg new file mode 100644 index 0000000..3b080f3 --- /dev/null +++ b/favicon.svg @@ -0,0 +1,4 @@ + diff --git a/free-audit.html b/free-audit.html new file mode 100644 index 0000000..cffa96a --- /dev/null +++ b/free-audit.html @@ -0,0 +1,201 @@ + + + + + + Free Audit — Krusty Planet + + + + + + + + +

Free AI Audit

No strings, no sales pitch, no catch

+ +

+ We'll spend up to 30 minutes looking at your business workflow and tell you + exactly where AI could save you time or money. You get a written report with + specific, actionable recommendations. If none of them make sense for you, that's + fine too. No follow-up calls, no obligation. +

+ +

What you get

+ +

🔍 Workflow Audit

+ You describe your daily operations. We map out repetitive tasks, bottlenecks, + and manual processes that could be automated. You'll get a prioritized list of + opportunities with estimated time savings. +

🔐 Privacy Check

+ If you're already using AI tools (ChatGPT, Copilot, etc.), we check whether + your data is being used for training, stored on third-party servers, or + exposed in ways your customers wouldn't expect. Most small businesses don't + know this is happening. +

🤖 Tool Fit Assessment

+ There are hundreds of AI tools. We cut through the noise and recommend + 2-3 that actually fit your specific use case, budget, and technical comfort + level. Includes self-hosted options that keep your data private. +

📊 Automation Opportunity Report

+ A written document (PDF or email) covering what we found, ranked by + impact and difficulty. Each recommendation includes an estimated cost + to implement and whether you could do it yourself with off-the-shelf + tools or need custom help. +

💬 Prompt Engineering Review

+ Already using AI but getting mediocre results? Send us your current prompts + and workflows. We'll rewrite them for better output, show you what changed, + and explain why. Often a 5-minute fix that saves hours. +

🏗️ AI Readiness Check

+ Thinking about deploying AI on your own hardware? We assess your current + infrastructure, bandwidth, and security posture. You'll know exactly what + you'd need to self-host AI agents and what it would cost. +

+ +

How it works

Fill out the form below with a brief description of your business and what you'd like audited.
We review within 48 hours. If we need more info, we'll ask one clarifying question.
You get a written report delivered by email. Plain English, no jargon, no sales pitch.
You decide whether any of it's worth pursuing. If yes, we can talk. If not, you still have the report.

+ +

The "no catch" part

No email list signup. No newsletter. No retargeting ads.
We don't add you to a CRM or follow up if you don't respond.
The report is yours to keep, share, or ignore.
Why free? These audits cost us almost nothing to produce, and they're the best way we know to demonstrate competence. If we can spot real opportunities, you'll know we're the real deal. If we can't, you'll know that too.

+ +

Request your audit

+ +

+ Name +

+ Email +

+ Business name +

+ What do you want audited? +

+ +

+ Describe your situation +

+ +

+ Your submission is processed via email and deleted from the server after delivery. + No newsletter, no CRM, no follow-up unless you respond. +

+ +

+ + + + + + diff --git a/index.html b/index.html new file mode 100644 index 0000000..0dcf959 --- /dev/null +++ b/index.html @@ -0,0 +1,148 @@ + + + + + + Krusty Planet — AI Consulting + + + + + + + + + + +

🌍 Krusty Planet

Self-hosted AI agents. Built on open-source, configured for your workflows.

+ Our Services + Get in Touch +

+ + +

What We Do

🤖 AI Agent Deployment

Set up OpenClaw, Ollama, or other open-source agent runtimes on your hardware. We handle configuration, model connections, and a working baseline you can extend.

🧠 Custom Skills

Build capabilities for your agents: email automation, data processing, web scraping, scheduled reports, API integrations. Each skill is tailored to a specific task you need handled.

🏗️ Agent Fleet Architecture

Design multi-agent systems where specialists collaborate: one agent researches, another writes code, a third audits the output. Task delegation with built-in quality checks.

🔐 Private Infrastructure

Everything runs on your hardware. Local model inference, no API calls to third parties, no data leaving your control. Optional air-gapped setups for sensitive environments.

🛠️ Ongoing Support

Monthly retainers for monitoring, health checks, skill updates, and troubleshooting. Critical issues addressed within 4 hours, standard requests within 24 hours.

🎓 Training & Documentation

Plain-English documentation for your team to operate and extend the system. Walkthrough sessions available so you're not dependent on us long-term.

+ + +

Privacy-first, not privacy-later.

+ We build on open-source tools (OpenClaw, Ollama, local LLMs) and deploy to your infrastructure. + Your models, your hardware, your data. We don't store your prompts, files, or agent outputs + on our systems. Contact form submissions are processed via email and deleted from the server + after delivery. +

🔒 Self-hosted by default
🌐 Open-source stack
📡 Available on I2P darknet
💰 Accepts Monero (XMR)
📝 Minimal logging policy

+ + +

Ready to get started?

Tell us what you're building and we'll figure out if we're a fit.

+ Start a Conversation +

+ + + + + + + diff --git a/js/main.js b/js/main.js new file mode 100644 index 0000000..e64e85d --- /dev/null +++ b/js/main.js @@ -0,0 +1,51 @@ +// Theme cycle: dark → light → forest → coffee +(function() { + const themes = ['dark', 'light', 'forest', 'coffee']; + const icons = { dark: '🌙', light: '☀️', forest: '🌿', coffee: '☕' }; + + function getStored() { + try { return localStorage.getItem('kp-theme'); } catch(e) { return null; } + } + + function setStored(theme) { + try { localStorage.setItem('kp-theme', theme); } catch(e) {} + } + + function applyTheme(theme) { + document.documentElement.setAttribute('data-theme', theme); + setStored(theme); + var btn = document.getElementById('theme-toggle'); + if (btn) btn.textContent = icons[theme] || '🌙'; + } + + // Init: stored preference, then data-theme attr, then default dark + var stored = getStored(); + var current = stored || document.documentElement.getAttribute('data-theme') || 'dark'; + if (themes.indexOf(current) === -1) current = 'dark'; + applyTheme(current); + + // Toggle + document.getElementById('theme-toggle').addEventListener('click', function() { + var idx = themes.indexOf(document.documentElement.getAttribute('data-theme')); + var next = themes[(idx < 0 ? 0 : idx + 1) % themes.length]; + applyTheme(next); + }); +})(); + +// Mobile nav toggle +document.getElementById('nav-toggle').addEventListener('click', function() { + var menu = document.getElementById('nav-menu'); + this.classList.toggle('is-active'); + menu.classList.toggle('is-active'); +}); + +// Contact form (placeholder) +var form = document.getElementById('contact-form'); +if (form) { + form.addEventListener('submit', function(e) { + e.preventDefault(); + var status = document.getElementById('form-status'); + status.style.display = 'block'; + form.querySelector('button[type="submit"]').disabled = true; + }); +} diff --git a/pricing.html b/pricing.html new file mode 100644 index 0000000..8448d21 --- /dev/null +++ b/pricing.html @@ -0,0 +1,139 @@ + + + + + + Pricing — Krusty Planet + + + + + + + +

Pricing

Fixed rates, concrete deliverables, no surprises.

+ +

+ + +

Hourly Rate

$50/hour

Applies to all services below. Billed in 15-minute increments.

+ + +

Fixed-Price Packages

+ +

Starter Kit

$150

One-time

Pre-configured OpenClaw or Ollama agent image
3 skills from our catalog (file summarization, web search, email drafting, code review, or one custom)
Step-by-step deployment guide written for your environment
Email support for 30 days (48-hour response)

Estimated work: 3 hours. Fixed price regardless of actual time.

+ Get Started +

+ +

Custom Build

$500

One-time

Up to 3 custom skills built for your specific workflows
Multi-agent setup (2-3 agents with task delegation)
Integration with one existing tool or API
Written documentation + 30-minute walkthrough call
Email support for 60 days (24-hour response)

Estimated work: 10 hours. Fixed price regardless of actual time. Extra work billed hourly at $50/hr.

+ Discuss Project +

+ +

Monthly Support

$200/month

Month-to-month, cancel anytime

Up to 2 hours of work per month (skill updates, troubleshooting, config changes)
Agent health monitoring and alerting
Critical issues responded to within 4 hours during business hours
Standard requests within 24 hours

Additional hours beyond 2/month billed at $50/hr. Business hours: Mon-Fri, 9am-5pm ET.

+ Set Up Support +

+ + +

Accessibility

+ If our rates don't work for your situation, ask about our needs-based sliding scale. + We also offer pro bono exemptions for qualifying nonprofit organizations. + Get in touch and we'll figure something out. +

+ + +

🌐 I2P Darknet

All services available via I2P eepsite for maximum privacy. Payments accepted in Monero (XMR). Contact us for our I2P address.

+ +

Accepted payment methods:

+ Stripe + PayPal + Monero (XMR) + Invoice +

+ + + + + + diff --git a/services.html b/services.html new file mode 100644 index 0000000..71d9b7a --- /dev/null +++ b/services.html @@ -0,0 +1,91 @@ + + + + + + Services — Krusty Planet + + + + + + + +

Services

What we build and how it works

+ +

🤖 AI Agent Deployment

Deploy AI agents on your hardware using OpenClaw, Ollama, or other open-source runtimes. We configure the environment, connect your preferred model (local or API), and hand you a documented, working system you can extend on your own.

OpenClawOllamaSelf-Hosted LLMsAPI Integration

+ +

🧠 Custom Skill Development

Each skill is built for a specific task in your workflow. Examples: an agent that reads your inbox and drafts categorized responses, a scraper that compiles daily reports from three websites, or a monitoring skill that alerts you when a service goes down.

Natural Language TasksAPI WorkflowsData ProcessingWeb Automation

+ +

🏗️ Agent Fleet Architecture

Multiple agents working in parallel, each with a specialty. A research agent finds sources, a coding agent implements, an audit agent reviews the output. Task delegation with built-in quality checks. We design the topology and write the coordination logic.

Multi-Agent OrchestrationTask DelegationQuality PipelinesParallel Processing

+ +

🔐 Private Infrastructure

Deploy on your own hardware, no third-party API calls required. We set up local model inference, encrypted communications, and optional air-gapped configurations for sensitive environments. Your data never touches our servers.

Air-Gapped DeployLocal InferenceE2E EncryptionI2P Hosting

+ +

🛠️ Ongoing Support

Monthly retainer for monitoring agent health, applying updates, building new skills, and troubleshooting issues. Critical problems addressed within 4 hours during business hours, standard requests within 24 hours.

Monthly RetainerHealth MonitoringSkill UpdatesTroubleshooting

+ +

🎓 Training & Documentation

Plain-English documentation for operating and extending your agent system. Walkthrough sessions (video call or text-based) so your team can maintain things independently. The goal is to make you self-sufficient, not dependent on us.

Written DocsWalkthrough SessionsTeam Onboarding

+ +

+ Discuss Your Project +

+ + + + + +