Particles

Name: Particles
Author: LinkedParticles

by LinkedParticlesversion 1.224.6i★ –iRuns on your computer

Registry nameorg.linkedparticles/particles

Install

Overview

Description from the maker (original English)

An append-only ledger of sourced, dated, confidence-scored claims your agent cites and you audit

PopNx MCP’s three pillars

These three are never merged into one score: opinions, measurements and popularity are different things.

User voiceiNo reviews yetReviews open together with accounts in a later step
Our basic checksi1 of 1 passedChecked from public information; code not scanned yet
UsageiNot measured yet★ –

What you can do with it

We have not written a plain-language summary for this one yet. Below are the facts we read from the registry.

Connection: Runs on your computer
Transport: stdio
Category: Accounting, finance & tax
You need to fill in: ANTHROPIC_API_KEY

Install

Pick the AI app you use, then copy the settings.

Before you install

Basic checks: 1 of 1 passed
Runs on your computer with your user’s permissions (files, network, apps).
You fill in yourself: ANTHROPIC_API_KEY, DATABASE_URL, PARTICLES_CONFIG

Values marked secret are keys to your account. Never share them or paste them into a chat.

You need

PyPIPython’s uv installed (for the uvx command)

Which app do you use?

ClaudeClaude CodeCursorVS CodeChatGPT

Pick your app above. We remember it, so next time the button says “Install in …”.

Open Claude Desktop → Settings → Developer → Edit Config, and paste this into claude_desktop_config.json

{
  "mcpServers": {
    "particles": {
      "command": "uvx",
      "args": [
        "linkedparticles",
        "mcp",
        "serve"
      ],
      "env": {
        "ANTHROPIC_API_KEY": "your-value",
        "DATABASE_URL": "your-value",
        "PARTICLES_CONFIG": "your-value"
      }
    }
  }
}

Run this in your terminal

claude mcp add particles --env ANTHROPIC_API_KEY=your-value --env DATABASE_URL=your-value --env PARTICLES_CONFIG=your-value -- uvx linkedparticles mcp serve

Fill in the empty values before running it.

Open in Cursor

Or copy the settings yourself

Paste into .cursor/mcp.json in your project, or ~/.cursor/mcp.json

{
  "mcpServers": {
    "particles": {
      "command": "uvx",
      "args": [
        "linkedparticles",
        "mcp",
        "serve"
      ],
      "env": {
        "ANTHROPIC_API_KEY": "your-value",
        "DATABASE_URL": "your-value",
        "PARTICLES_CONFIG": "your-value"
      }
    }
  }
}

Your browser asks to open the app, then the app shows the settings and asks you to confirm. Key values are left empty for you to fill in.

Open in VS Code

Or copy the settings yourself

Paste into .vscode/mcp.json in your project

{
  "servers": {
    "particles": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "linkedparticles",
        "mcp",
        "serve"
      ],
      "env": {
        "ANTHROPIC_API_KEY": "your-value",
        "DATABASE_URL": "your-value",
        "PARTICLES_CONFIG": "your-value"
      }
    }
  }
}

Your browser asks to open the app, then the app shows the settings and asks you to confirm. Key values are left empty for you to fill in.

This one runs on your computer, and ChatGPT can only connect to servers on the internet. Pick another app.

Your choice stays on this device in a cookie; no account needed.

Values you fill in

ANTHROPIC_API_KEY secret required env
Anthropic API key. Required for the tools that call a model — `query` composes a cited natural-language answer, and the semantic half of `lint` reasons over claim pairs. The accessors that only read…
DATABASE_URL env
SQLAlchemy URL of the belief store to serve, used as given, so write a SQLite path absolute: a relative one opens against the directory the server starts in. Without it, and without PARTICLES_CONFIG,…
PARTICLES_CONFIG env
Path to the config file holding the rest of the server's settings: additional stores, model selection per purpose, and the `mcp.write.enabled_stores` allowlist that turns the write tools on. This is…

These settings are generated from the maker’s registry entry. If the README says otherwise, follow the README.

What it can do

Runs on your computer; the tool list appears in your AI app after you install it.

Basic checksi

We check information anyone can see; we have not read the code. Checks that do not apply are left out of the count. · Checked Nov 6, 2026

✓ Pass
Maker matches the repo
The registry name was verified against the domain linkedparticles.org (the registry checks domain ownership)
… Not checked yet
Still maintained
We have not read this repo from GitHub yet
… Not checked yet
Has a licence
We have not read this repo from GitHub yet
– Not applicable
Endpoint responds
This one runs on your computer; there is no internet endpoint
– Not applicable
Sign-in with OAuth
Runs on your computer and uses local settings instead
… Not checked yet
No hidden instructions in tool descriptions
Runs on your computer; we can only read tool lists from internet endpoints

Compared with others in Accounting, finance & taxi

More stars than 0% of the other 1720 servers in this category
Published to the registry more recently than 88% of the other 1720 servers in this category
Basic checks equal to or better than 100% of the other 1720 servers in this category

Compared with every active server in this category in the registry. Stars count only when the publisher owns the repo.

Common questions

Is Particles safe?

It passed 1 of 1 applicable basic checks. We have not read the code. Before adding a key or token, see what access the server asks for and use a key with only the permissions you need.

What keys or codes does it need?

You need to fill in: ANTHROPIC_API_KEY

Does it run on my computer or in the cloud?

It runs on your computer. Your AI app starts it when needed, and data does not pass through the maker’s servers (unless it calls an online service itself).

Is it still maintained?

The GitHub repo last changed –, and the latest version (1.224.6) reached the registry 4 days ago.

Which apps can use it?

Apps that support MCP, such as Claude Desktop, Claude Code, Cursor and VS Code. You also need Python (uv) on your computer.

What can I use instead?

In the same category: Finance Toolkit, SolidInvoice, Korea Tax Calculator

Alternatives

Servers in the same category from other makers, sorted by stars.

Reading a card:★ GitHub starsiThree pillars: reviews · checks · usageiUpdated = latest version published to the registryi

Finance Toolkit

by JerBouma

500+ transparent financial methods calculated from raw statements, not third-party endpoints.

Computer or cloud★ 5,496MITUpdated yesterday

Install

SolidInvoice

by SolidInvoice

Create and send invoices and quotes, manage clients, and track payments in SolidInvoice.

Runs in the cloud★ 981MITUpdated 7 days ago

Install

Korea Tax Calculator

by UpstageAI

Korean tax calculation and tax-audit judgment MCP: 95 tools with statutory citations.

Runs on your computer★ 97MITUpdated today

Install

README from the maker

Open the README

Copied from the maker’s repo as written. PopNx MCP has not reviewed or endorsed it. HTML, images and scripts are removed. Links lead off this site.

linkedparticles

[linkedparticles.org](https://linkedparticles.org) · [docs.linkedparticles.org](https://docs.linkedparticles.org) · PyPI · Apache-2.0

Particles is shared memory for humans and AI agents. Each particle is one

claim, plus what you need to judge it: who said it, where, when, and how

confident they were. Facts, opinions, and memories are all claims, recorded

the same way as particles. Particles are not edited or deleted. Particles are

superseded, retracted, or disputed in the open. How much to trust it is a

perspective applied at query time, never baked into the record.

linkedparticles is the Engine of the Particles reference implementation: the package you install to actually run a belief store. Give it documents, pages, chat logs, or an agent's own session notes; get back claim-granularity beliefs, each carrying calibrated confidence, an uncertainty kind, a resolved subject, and provenance back to the exact source bytes. Ask a question and the answer cites the beliefs it was built from. Nothing is ever overwritten, so you can ask the store what it believed a year ago and why it stopped.

When your agent is wrong, you can see exactly why, and fix it at the source.

The Engine is a library first. The HTTP API, the CLI, the read-only MCP server, and the resident daemon are all *surfaces* over it.

Install

pip install linkedparticles

Python 3.11+. linkedparticles-core, the store-free Client layer, is pulled in automatically.

Extraction and the written answers call a language model. The quickstart below uses Anthropic, and any OpenAI-compatible endpoint works too, hosted or local (Ollama, llama.cpp, vLLM): each purpose names its provider in config.yaml (named providers). Deposit, the session digest and structural queries make no model call at all.

Sixty seconds

export ANTHROPIC_API_KEY=sk-ant-...

particles db init

Deposit a source (file, URL, or literal text) into the append-only corpus

particles deposit https://en.wikipedia.org/wiki/Pluto

Extract claim-granularity beliefs with confidence, subjects, and provenance

particles extract --all-pending

Ask in natural language; the answer cites the particles behind it

particles query "Why is Pluto not a planet?"

Health-check the store: contradictions, staleness, orphaned links

particles lint

Because nothing is overwritten, the store can replay its own history. --as-of is the assertion-time lens. It answers *what did the store believe at T, and why did it stop*, not what was true of the world at T:

particles query "How many planets are in the Solar System?" --as-of 2000-01-01

→ the belief as it stood then, and why it no longer stands: SUPERSEDED,

retired 2006-08-24, superseded by "Pluto is a dwarf planet."

There is a clickable version of that belief history on the front door. Step by step: getting started, then as-of time travel.

Wiring it into an agent

particles mcp serve                 # read-only MCP server (stdio), any MCP client

particles init claude-code # session-start digest, session-end harvest particles engine serve # HTTP API, for a shared or remote engine particles export obsidian ./vault # also: anki, wiki, logseq, notion, graph, jsonl

It can also stand in for the reference memory server, and it plugs into LangChain as a retriever and a tool set. See Claude Code memory, swapping in for the reference memory server, and using from LangChain.

Does it hold up?

Measured on LongMemEval, the multi-session agent-memory benchmark, with a full-context oracle and a no-memory floor run as controls in the same harness:

	Recall@40	End-to-end QA
Particles memory (shipped default, `top_k` 40)	0.969	0.804
Full-context oracle (baseline, n=148)	—	0.878
No memory (floor)	—	0.047

The oracle and floor rows come from a separate control run in the same harness, the day before the three runs the Particles row averages; the benchmarks page names the report of record for each row.

Method, per-question-type breakdowns, comparator memories, and the budget-matched arm (including the arms where Particles *loses*) are on the benchmarks page, with the report JSONs of record beside them.

Security posture

A memory layer is an *input-shaping* attack surface: it does not merely hold data, it shapes what the agent believes and does next, so a poisoned claim is an instruction on a delay timer.

Fifteen adversarial audits have covered this code, and each returned GO-WITH-FIXES. The verdicts are quoted as given.

Audit	Code audited	Verdict	Fixes shipped in
June 2026, before the code was opened: the HTTP, CLI and MCP surfaces, the filesystem writers, the egress layer, the CI and wheel build	`v1.62.0`	GO-WITH-FIXES	`1.62.1` to `1.63.0`; the last open finding closed in `1.128.0`
2026-10-23: everything changed since `v1.62.0`, including the release and publication pipeline the first audit predates, and the claims on this page	`v1.210.2`	GO-WITH-FIXES	`1.210.3` to `1.211.0`
2026-10-24: everything changed since the second audit, every earlier finding, and the claims on this page again	`v1.211.0`	GO-WITH-FIXES	`1.211.1` to `1.214.0`
2026-10-24, the third delta audit: everything changed since the previous audit, every earlier finding still open, and 92 checks of the claims on this page and the pages it links	`v1.214.2`	GO-WITH-FIXES, with five Medium findings and 22 claims that did not hold	`1.214.3` to `1.216.0`; claims restated in `1.216.1`
2026-10-25, the fourth delta audit: everything changed since the third delta audit, every earlier finding still open, and 125 checks of the claims on this page and the pages it links	`v1.216.1`	GO-WITH-FIXES, with nine Medium findings, one regression and 29 claims that did not hold	`1.217.1` to `1.220.1`; claims restated in `1.219.2`
2026-10-25, the fifth delta audit: everything changed since the fourth delta audit, every earlier finding still open, and 151 checks of the claims on this page and the pages it links	`v1.220.1`	GO-WITH-FIXES, with three Medium findings and 9 claims that did not hold	`1.220.2` to `1.220.9`; claims restated in `1.220.12`
2026-10-26, the sixth delta audit: everything changed since the fifth delta audit, per-client engine keys among it, every earlier finding still open, and 233 checks of the claims on this page and the pages it links	`v1.221.1`	GO-WITH-FIXES, with seven Medium findings and 53 claim judgements that did not hold	`1.221.2` to `1.221.5`; claims restated in `1.221.6`
2026-10-27, the seventh delta audit: the sixth audit's fixes, every earlier finding still open, and 373 judgements of the claims on this page and the pages it links	`v1.221.6`	GO-WITH-FIXES, with six Medium findings, one regression and 46 claim judgements that did not hold	`1.221.7` to `1.221.9`; claims restated in `1.221.10`
2026-10-28, the eighth delta audit: the seventh audit's fixes, the earlier findings still open, and 131 checks of the claims on this page and the pages it links	`v1.221.10`	GO-WITH-FIXES, with two Medium findings and 31 claim checks that did not hold	`1.221.11` to `1.222.0`; claims restated in `1.222.1`
2026-10-28, the ninth delta audit: the eighth audit's fixes, the earlier findings still open, and 226 checks of the claims on this page and the pages it links	`v1.222.1`	GO-WITH-FIXES, with three Medium findings and 52 claim checks that did not hold	`1.222.2` to `1.223.0`; claims restated in `1.223.1`
2026-10-30, the tenth delta audit: the ninth audit's fixes, the earlier findings still open, and 224 checks of the claims on this page and the pages it links	`v1.223.1`	GO-WITH-FIXES, with one Medium finding and 45 claim checks that did not hold	`1.223.2`; claims restated in `1.223.4`
2026-10-30, the eleventh delta audit: the tenth audit's fixes, the earlier findings still open, and 234 checks of the claims on this page and the pages it links	`v1.223.4`	GO-WITH-FIXES, with four Medium findings and 65 claim checks that did not hold	`1.223.5`; claims restated in `1.223.6`
2026-10-31, the twelfth delta audit: the eleventh audit's fixes and the claims on this page and the pages it links, with the earlier findings re-checked where their code changed	`v1.223.6`	GO-WITH-FIXES, with five Medium findings and 13 of 69 claim rows that did not hold	`1.223.7`; claims restated in `1.223.8`
2026-10-31, the thirteenth delta audit: the twelfth audit's fixes and the claims on this page and the pages it links, with the earlier findings re-checked where their code changed	`v1.223.8`	GO-WITH-FIXES, with four Medium findings, one regression and 8 of 69 claim rows that did not hold	`1.223.9`; claims restated in `1.223.10`
2026-10-31, the fourteenth and last delta audit: the thirteenth audit's fixes and the claims on this page and the pages it links, with the earlier findings re-checked where their code changed	`v1.223.10`	GO-WITH-FIXES, with four Medium findings and 7 of 69 claim rows that did not hold	`1.223.11`; claims restated in `1.223.12`

Not every release in those ranges contains the releases numbered before it: 1.212.0, 1.214.8 and 1.215.0 were tagged before their merge. Every release from 1.215.1 on contains every lower release. Nor does a release number prove its commit passed CI: 1.222.3, 1.222.4 and 1.222.5 were tagged on commits whose CI run failed, and the release tagger now tags a version only after its CI run passes.

Three of the fix rounds changed the design rather than a defect. An engine with no key serves reads only: it refuses every write, from loopback too, and the particles command line, the Claude Code hooks and the MCP servers write the local store in-process and need no key. Each client can hold a key of its own whose role bounds it: a reader key writes nothing, an agent key writes only as its own principal, and PARTICLES_API_KEY and every per-client operator key carry the operator's authority, by design, so give an operator key only to clients you would give that authority. The local refresh trusts a store, its config file or a rule file only when you alone can write it.

A cloned repository cannot configure Particles, or give a stdio MCP server or a Claude Code hook its store, unless a relative path you set points into it (a relative PARTICLES_CONFIG for the command line, a relative DATABASE_URL for an MCP server) or particles init claude-code --store default runs inside it. A store it ships, found by a path relative to the directory, cannot direct a refresh to a file on your disk, and that store's blob directory cannot direct a blob read or write to another file. No HTTP route reads a file the caller names on the engine's disk. A web page cannot send writes to an engine on loopback, or read from one that runs without a key, and

Only the beginning is shown; read the rest in the repo

Particles

Overview

PopNx MCP’s three pillars

What you can do with it

Install

Before you install

Values you fill in

What it can do

Basic checksi

Maker matches the repo

Still maintained

Has a licence

Endpoint responds

Sign-in with OAuth

No hidden instructions in tool descriptions

Compared with others in Accounting, finance & taxi

Common questions

Alternatives

Finance Toolkit

SolidInvoice

Korea Tax Calculator

README from the maker

linkedparticles

Install

Sixty seconds

Deposit a source (file, URL, or literal text) into the append-only corpus

Extract claim-granularity beliefs with confidence, subjects, and provenance

Ask in natural language; the answer cites the particles behind it

Health-check the store: contradictions, staleness, orphaned links

→ the belief as it stood then, and why it no longer stands: SUPERSEDED,

retired 2006-08-24, superseded by "Pluto is a dwarf planet."

Wiring it into an agent

Does it hold up?

Security posture