Get This Tool
Alchimist
Pricing
- Model
- Free
Summary
Generic blank-box AI prompting produces generic output — because without context, the model has no anchor. ΔLCHIMIST wires a browser extension directly to the page you're reading, so every generation cycle runs against real, live content rather than whatever you managed to paste into a chat window.
The extension parses the active web page and feeds that content as grounded context into your chosen persona, then generates output without sending anything to a hosted backend — your API key, your browser, your data. Persona-based rewriting lets you reshape or refine page content through a defined voice, which matters when you're producing client work that needs to sound like someone specific. The client-side architecture is the core differentiator: nothing leaves the browser except the call you make to your own API endpoint. That same architecture is the ceiling — there is no shared workspace, no team queue, and no way to chain steps across sessions without starting over.
Bottom line: Pick this for solo ghostwriting or research workflows where you need page-anchored output and cannot accept third-party data exposure — but if your workflow requires multiple collaborators or multi-step automation across sessions, the single-user, one-shot model breaks immediately.
Community Performance Report Card
No community ratings yet. Be the first to rate this tool!
Community Benchmarks Community
Sign in to submit a benchmarkNo community benchmarks yet. Be the first to share a real-world data point.
Pros
Sign in to edit- Full client-side execution with BYOK, so no content ever reaches a vendor-controlled server — which means teams handling confidential client copy or sensitive research data get a hard architectural guarantee, not a privacy policy promise.
- Page content is parsed and used as the grounded context for every generation cycle, so output is anchored to what you are actually reading rather than what you remembered to paste — which eliminates the copy-paste-and-hope workflow that produces off-target drafts.
- Persona configuration shapes output voice consistently across rewriting and refinement passes, so ghostwriters producing work for a specific client voice do not have to re-specify tone on every prompt.
- Multi-stage interaction with page arguments is described in the repo, so a researcher can push back on a source's claims across several passes without leaving the extension or manually re-feeding context each time.
Cons
Sign in to edit- There is no collaboration layer or shared workspace. The moment a second person needs to review, edit, or continue a session, they are starting from scratch — teams doing any kind of coordinated content production will immediately hit this wall and route around it by exporting output manually, which is a workflow, not a feature.
- Session state does not persist across browser sessions. Any multi-stage work in progress is lost when the tab closes or the browser restarts, which means long research threads or iterative drafting projects require manual continuity management by the user.
- The license is custom and restrictive per curation review — teams intending to fork the extension, embed it in a product, or redistribute a modified version face legal ambiguity that a standard MIT or Apache-licensed tool would not create. When that matters, teams switch to an extension built on a clearly permissive license.
- There is no API surface and no integration hooks. Output cannot be pushed to a CMS, a project management tool, or any downstream system without manual copy-paste — at the point where a team needs even basic automation around the output, this tool is not part of the answer.
Community Reviews
Sign in to write a reviewNo reviews yet. Be the first to share your experience.
About
- Platforms
- Browser (Chrome extension)
- API Available
- No
- Self-Hosted
- Yes
- Last Updated
- 2026-06-25T18:18:25.897Z
Best For
Who it's for
- Professional creators and ghostwriters
- Researchers needing page-anchored output
- Users wanting client-side AI without data exposure
What it does well
- Context-grounded content generation on any web page
- Persona-based rewriting and refinement of page content
- Multi-stage strategic interactions with page arguments
Discussion Community
Sign in to commentNo discussion yet. Sign in to start the conversation.
Compare Alchimist
Spotted incorrect or missing data? Join our community of contributors.
Sign Up to ContributeCommunity Notes & Tips Community
Sign in to contributeBe the first to contribute. General notes, observations, gotchas, and tips from people who use this tool day-to-day.
Frequently Asked Questions
- Is Alchimist free?
- Yes — Alchimist is fully free to use. There is no paid tier.
- Is Alchimist open source?
- Yes. Alchimist is open source.
- Can I self-host Alchimist?
- Yes. Alchimist supports self-hosting on your own infrastructure.
- What platforms does Alchimist support?
- Alchimist is available on: Browser (Chrome extension).
Hours Saved & ROI Stories Community
Sign in to contributeBe the first to contribute. Concrete time/cost savings, with context. e.g. "Cut my code review backlog from 4h to 45m per week."
Curated lists that include this category
ΔLCHIMIST is a fully client-side browser extension that converts any active web page into a grounded authoring context. The extension reads and parses the page you have open, reflects that content back through a persona you configure, and produces output tied to what is actually on the page — not a generic prompt submitted in a vacuum. The core workflow is: open a page, select or define a persona, trigger a generation cycle, and receive output that the vendor describes as anchored to the live content. All execution happens inside the browser’s runtime using keys you supply (BYOK), so no content transits a hosted service.
The persona system is the differentiating feature. Rather than a single system prompt, the extension is built around what the repo describes as ‘persona alchemy’ — the ability to define a specific authoring voice and apply it consistently across rewriting, refinement, and multi-stage interactions with the page’s arguments. For ghostwriters producing work that must sound like a particular client, or researchers who need output that reflects a defined analytical stance, this is the mechanism that keeps output from drifting toward generic model defaults.
The tool fits solo professionals who work page-by-page: ghostwriters drafting around source material, researchers synthesizing a live article, or operators who handle sensitive content and need a hard guarantee that data stays local. It does not fit team workflows — the architecture has no collaboration layer, no shared session state, and no API surface for integrating output into a downstream pipeline. Multi-stage interactions are described in the repo, but each session is self-contained; there is no persistent memory or cross-session chaining. Teams that outgrow single-user, single-session constraints will find themselves manually stitching output together or moving to a platform with a proper pipeline model.
The extension installs as a standard browser add-on. API keys are stored and used within the browser’s execution environment. The repo references a LICENSE file, but the license terms observed during curation are custom and restrictive rather than a standard open-source permissive license — teams intending to fork or redistribute should read those terms before building on this codebase.