HoneClaw: A Rust-Based Investment Research Assistant That Argues With You
HoneClaw (Hone-Financial) is dedicated to being a professional investment assistant that truly understands you.
At a glance
- What is it?
- HoneClaw is an open-source, Rust-built personal investment assistant that tracks companies, enforces discipline, and integrates with chat platforms. It is designed to challenge your reasoning rather than agree with you.
- Who is it for?
- Adopt HoneClaw if you are an individual investor who wants a self-hosted, chat-driven research tool that enforces discipline and keeps long-term memory. Skip it if you need a turnkey product or prefer plain text workflows without a Rust setup.
- Can I use it commercially?
- Yes. MIT is a permissive licence: you can use, modify and sell software built on it, as long as you keep its copyright and licence notices.
- Is it still maintained?
- Yes. The repository last received commits 2 days ago.
- What is it written in?
- Mainly Rust, according to GitHub's language statistics.
Answers come from the project's GitHub data, last synced on September 14, 2026, and from our analysis. They are not legal advice.
DEEP OPEN-SOURCE ANALYSIS
What HoneClaw Actually Fixes
Most investment chatbots agree with you. HoneClaw is built to do the opposite. The README states it is "not a chat toy designed to indulge you, but a ruthless defender of your investment discipline." The problem it solves is emotional trading: chasing news, reacting to price swings, and abandoning a pre-set plan. It targets individual investors who want a co-pilot that cross-checks decisions against data and predefined rules. The core value is not prediction. It is restraint. The tool tracks companies you hold, maintains long-term profiles in Markdown, and alerts you when your own take-profit or stop-loss conditions trigger. That is a specific workflow, not a general-purpose assistant.
How the Mechanism Works
The architecture is visible in the repository layout and the interactive HTML diagram. The core is Rust, using Tokio, Axum, and SSE for the backend. The client is a Tauri-based desktop app, and the frontend uses SolidJS with Tailwind v4. The data flow starts with messages from supported channels: Web, iMessage, Lark, Telegram, and Discord. Those messages hit the Rust backend, which processes them against stored company portraits and event timelines. Long-term memory is kept as Markdown files, which is a deliberate choice. It means your research history is readable and portable. Scheduled tasks run as cron jobs, triggering pre-market briefings, post-market summaries, or post-earnings analysis. The system then pushes alerts back through the same chat channels. The key mechanism is the discipline engine: it compares your stated logic and levels against market data, then flags flaws in your reasoning.
Three Ways to Get It Running
The README offers three installation paths. The one-line install fetches a shell script and runs it: curl -fsSL https://raw.githubusercontent.com/B-M-Capital-Research/honeclaw/main/scripts/install_hone_cli.sh | bash. After that you run hone-cli doctor, then hone-cli onboard, then hone-cli start. In a second terminal you launch the web interfaces with hone-cli web admin-ui and hone-cli web user-ui. The Homebrew route is similar: brew install B-M-Capital-Research/honeclaw/honeclaw, then the same hone-cli commands. For development, you clone the repo and run cargo run -p hone-cli -- start --build, plus separate terminals for the admin and user UIs with the --dev flag. All three paths assume a Unix-like system, macOS or Ubuntu, and a stable Rust toolchain with Edition 2024 support. The wiki holds the full startup matrix, port numbers, and configuration keys, which the README points to but does not reproduce.
The Trade-Offs in the Design
The Markdown-based memory is a strength and a limitation. It keeps your research portable and diffable, but it is not a structured database. If you want to query across hundreds of companies by metric or date, you will be parsing files yourself. The README also admits that professional valuation tools, research workflows, and proprietary knowledge bases are not in the public repository. That means the open-source version is a shell. The discipline logic exists, but the deeper analytical models are missing. Another trade-off: the system is chat-first. If you do not live in Telegram or Discord, the value drops. The web console exists, but the emphasis is on IM integration. Finally, the performance claim is millisecond-level responsiveness. That is plausible for a Rust backend, but the README gives no benchmark numbers. Treat it as a design goal, not a measured fact.
Where It Is the Wrong Tool
HoneClaw is not for you if you want a fully managed service. There is no hosted option in the README; you must run it yourself. That means maintaining a Rust toolchain, handling the cron scheduler, and managing the Markdown files. If you are a casual investor who checks a portfolio once a month, the setup cost outweighs the benefit. It is also the wrong tool if you need real-time market data feeds out of the box. The README does not specify which data sources are used. You will likely need to configure your own. And if you prefer a simple spreadsheet or a plain text journal, the chat integration and cron jobs are overkill. The tool is opinionated. It assumes you want to be challenged. If you want a passive news aggregator, this is not it.
A Real Alternative: Plain Markdown Plus Cron
The closest alternative is not another chatbot. It is the thing HoneClaw is built on: a folder of Markdown files plus a cron job. You can write company notes by hand, set up a daily reminder with cron, and use a shell script to pull quotes from any public API. The difference in approach is control and structure. With raw Markdown, you own the format and the workflow. There is no discipline engine, so you must enforce your own rules. HoneClaw adds the chat interface, the monitoring logic, and the multi-channel alerts. But the underlying data is still Markdown. If you are comfortable with a text editor and a terminal, you can replicate the memory part in an afternoon. What you cannot replicate easily is the cross-platform chat layer and the scheduled task orchestration. That is the real value HoneClaw adds. The alternative is simpler, but it puts all the discipline burden on you.
Maintenance and License Reality
The project is under the MIT license, which is permissive. You can fork it, modify it, and use it commercially without paying. But the README notes that some components are not included. That is a maintenance risk: the public repo may lag behind the full product. The last push is recent, v0.15.3, and the release cadence suggests active development. However, active development is not a promise of stability. The wiki is the source of truth for ports, config keys, and troubleshooting. Expect to read it. Upgrading means pulling new releases and re-running hone-cli doctor to check your environment. The Markdown memory files are likely stable across versions, but the config format for IM channels and cron jobs may change. Budget time for reading release notes. The MIT license is friendly, but the missing pieces mean you are adopting a partial system. That is a concrete boundary to verify before you rely on it.
Editorial conclusion
Adopt HoneClaw if you are an individual investor who wants a self-hosted, chat-driven research tool that enforces discipline and keeps long-term memory. Skip it if you need a turnkey product or prefer plain text workflows without a Rust setup. Before adopting, verify the exact configuration keys for your IM channels and cron jobs in the wiki, and confirm which proprietary valuation tools are excluded from the public repo.
Community notes