SiteList review score
Enforces strong security headers and provides a clean llms.txt file. Robots.txt rules currently block Next.js rendering assets and suppress search indexation.
Reviewed on Aug 26, 2026
dropstone.io
Unclaimed listingWhat it does
Self-hosted AI assistant and coding agent that runs on user infrastructure, featuring long-term memory, proactive monitoring, smart home control, phone calls, and code generation across CLI, chat, and SDK surfaces.
SiteList review score
Enforces strong security headers and provides a clean llms.txt file. Robots.txt rules currently block Next.js rendering assets and suppress search indexation.
Reviewed on Aug 26, 2026
Verified user rating
—
No verified reviews yet — the evidence locker is empty.
Pro is $15/mo billed monthly; Max starts at $75/mo billed monthly. Annual billing saves 17%. Billed monthly.
| Plan | Price | Details |
|---|---|---|
| Free | $0 monthly per null | Everything you need to start shipping with Dropstone Limited monthly usage · Access to Dropstone CLI · Access to Dashboard · Generate code · Web search inside the agent · Memory across conversations · Connect any tool through Remote MCP · Dropstone Fast with XHigh Thinking · Dropstone Pro with XHigh Thinking · Standard output limits per task · Limited monthly usage credits · Standard concurrent tasks, 100MB file upload limit, 24-hour context window, 500 tokens/session, 30-day memory retention, 100 requests/day, 10GB storage, 1 user, View source |
| Pro | $15 monthly per null | More usage, the Heavy tier, and the research roadmap 6× more usage than Free · Access to Blankline Research · Latest engineering research upgrades · Access to Beta Research Program · First look at monthly audit results · Smart home device control · Dropstone Pro with XHigh Thinking · 8× Free usage credits · Standard output limits per task · Standard concurrent tasks, 100MB file upload limit, 24-hour context window, 500 tokens/session, 30-day memory retention, 100 requests/day, 10GB storage, 1 user, View source |
| Max | $75 monthly per null | The most usage, priority queue, and the earliest features 5× or 10× more usage than Pro · Priority routing during high-traffic windows · Dedicated capacity reserved at peak hours · Faster first-token latency on every tier · Early access to advanced Dropstone features · Preview new models before public release · Experimental agent presets, internal-only releases · Real phone calls placed on your behalf · Early access to new Real-World AI capabilities · Dropstone Heavy with XHigh Thinking · 5× or 10× Pro usage credits · Higher output limits per task, Higher concurrent tasks, 100MB file upload limit, 24-hour context window, 500 tokens/session, 30-day memory retention, 1000+ GB/m View source |
Long-term memory
Memory
Persists user preferences, context, and skills across sessions, surfaces, and tools without requiring re-prompting.
Proactive monitoring
Automation
Watches designated inputs or environments in the background and briefs the user when action is needed.
Real-world actions
Real-World AI
Places phone calls, controls smart home devices, monitors inboxes, and conducts deep research with citations.
Code generation & shipping
Engineering
Writes, reviews, and ships software via CLI, IDE, or SDK on user infrastructure with zero data retention on managed inference.
Team software development
Developers, founders, and designers collaborate in a shared workspace where AI handles implementation while humans review code and business requirements.
Real-world personal assistance
Automates inbox monitoring, phone calls, smart home control, and deep research for individual users who want proactive background support.
SmartThings
automation
Connects to see and control devices like TVs, lights, thermostats, locks, vacuums, and sensors via rules or manual commands.
Remote MCP
developer tools
Protocol for connecting any external tool, service, or API to the agent.
Dropstone is a developer-focused local AI agent runtime designed to manage codebases and home automation.
Dropstone scored 36/100 primarily due to technical crawl blockers in robots.txt that prevent search engines from rendering Next.js assets, combined with host fragmentation and missing JSON-LD schema.
Dropstone must unblock `/_next/` in robots.txt, enforce a 301 redirect between www and non-www domains, align homepage hero copy with search intent, and inject structured data.
Dropstone features an exceptional `/llms.txt` file and open crawler access, but missing server-rendered HTML text limits full AI extraction of main landing page messaging.
Answers are based on SiteList's latest evidence review. Read the full review.
Dropstone scores 32/100 on SiteList due to severe technical indexation blockers and rendering suppression in robots.txt. While its technical security posture and llms.txt implementation are strong, blocked Next.js assets and missing structured data critically limit search visibility.
32/100: strong security foundation, blocked by robots.txt
No reviews yet. The evidence locker is empty.
Reviews from verified users appear here once approved.
Reviews are from the SiteList community. Some reviewers received incentives — disclosed in the review, always. The verdict is never for sale.
Home
CodeBurn — See where your AI coding spend actually goes screenshot 1 of 4: Home
codeburn.app
CodeBurn is a free, open-source, local-first tool that tracks AI coding token usage and cost across dozens of developer tools. It breaks down spend by model, project, and task using session files already on the user's disk, with no data uploaded.
whop.sh
The Whop CLI gives developers, creators, and automation agents command-line access to Whop API resources.
Home
Shieldstral — Define safety at runtime for text and images screenshot 1 of 4: Home
mistral.ai
Shieldstral introduces a 3B open-weights multimodal safety classifier that outperforms models up to 7x its size.
Home
Soup CLI — Fine-tune an 8B LLM on a 4 GB laptop GPU screenshot 1 of 4: Home
trysoup.dev
Soup CLI is an open-source command-line toolkit for LLM post-training and fine-tuning on local hardware. It uses layer streaming to fit 8B models into 4 GB VRAM, auto-generates training configs, gates saves via evals, and supports migration from other fine-tuning tools.