SkillAgentSearch skills...

proxenos

A free, open-source agent that keeps your own records private on your computer and drafts the next step for you. Built first for people getting through the health care system.

Install / Use

claude mcp add mindemory -- npx -y github:mindemory/proxenos

If the server publishes to npm under a different name, use that package instead — check the repo README.

About this skill
🔌

MCP Server

Model Context Protocol server

Quality Score

83/100

Supported Platforms

Claude Code
Claude Desktop

Our assessment of proxenos

proxenos scores 83/100 on our quality scale, 3030th of 4,652 Development & Engineering skills we index.

Its MCP Server is 29 KB long, well organised into 10 sections with 7 code examples: a thorough specification that gives an agent plenty to work with.

It has 3 GitHub stars, so there is little community track record yet; judge it on its content.

Substance
30/30
Structure
20/20
Description
15/15
Adoption
3/20
Freshness
15/15

Maintenance, license and trust

  • The repository was last updated today, so proxenos is actively maintained.
  • It is released under the MIT license, a permissive license that allows use, modification and commercial use with attribution.
  • Its trust signals score 92/100, with 1 caution from licensing, adoption, age or documentation. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.

proxenos compared with similar skills

All 4 of these similar skills score higher than proxenos; compare them before choosing.

SkillScoreStarsUpdatedFormat
proxenos (this skill)by mindemory833todayMCP Server
Agent-Reachby Panniantong100107.4ktodayCLAUDE.md
taste-skillby Leonxlnx100107.4k1d agoCLAUDE.md
headroomby headroomlabs-ai10078.5ktodayCLAUDE.md
ai-job-searchby ai-job-search-dev10068.6k1d agoCLAUDE.md

Frequently asked questions

How do I install proxenos?
Run claude mcp add mindemory -- npx -y github:mindemory/proxenos. The install tabs above show the steps for each supported agent.
Which AI agents does proxenos work with?
It is written for Claude Code and Claude Desktop, as a MCP Server file. Other agents that read the same format can often use it too.
Is proxenos safe to use?
It is MIT-licensed and scores 92/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
Is proxenos still maintained?
The repository was last updated today, so proxenos is actively maintained.

Proxenos

checks macOS Python 3.12 MIT license Local-first GitHub stars XERJ Corpus Hub contributor

A free, open-source agent that keeps your own records private on your computer and drafts the next step for you. Built first for people getting through the health care system.

When you're unwell, the hard part is often everything around the care: referrals that go quiet, portals, bills that don't match, forms, and the same story told again at every visit. Proxenos reads what you already have (your mail, chats, calendar and notes), keeps track of where things stand, and writes the message, the call script or the one-page brief for your next appointment, in your own voice. You stay in charge: it drafts, and you press send.

Home: today's agenda and open tasks, drafts to send, and where health care stands The home page of proxenos demo, a made-up person with made-up records. More below.

It grew out of one person's weeks of getting through the system while unwell. The story is Try anything, but don't fall sick, and more on the project is at mindemory.io/proxenos.

Status: early, and used every day by one person. It runs on a Mac (Apple Silicon) for someone comfortable with a terminal, and every check passes from a fresh clone. It's about 59,000 lines of Python, TypeScript, Swift and JavaScript, with 23,000 more in 790 tests that run on macOS and Windows on every push. It isn't yet something you can download and use without code. That's the next track (docs/everyone.md): the privacy screen is built, and so is a first desktop app for Mac and Windows, unsigned for now; signing it, and signing in instead of configuring, come next. Expect rough edges, and readers that assume a particular export format.

Help of any kind is welcome: code, an issue, or an idea over a message or a call. Healthcare comes first, so people with medical training are especially welcome. Knowing what's worth tracking and what to ask is what would make this preventive, not just a way through the system. Good places to start are in CONTRIBUTING.md. If you'd like Proxenos to exist for more people, a star helps others find it.

Not medical advice. Proxenos organizes your own records and drafts messages. It doesn't diagnose or recommend treatment: the care view restates what your records say, and anything that matters belongs with your care team.

Proxenos gathers your own data (messages, email, calendar, notes, documents, activity) into one private vault on your computer. From that it keeps a written model of who you are, and runs small agents ("clones") that work for you:

  • a feed of new papers and news, ranked for you;
  • your open tasks, pulled from your conversations and from what you tell it, kept up through the day, with the plans among them (a lunch at 12, a call at 9) put on your own calendar, and taken off it when they're called off;
  • a journal of each day, so you never have to narrate it: what got done, where you wrote, what moved in your projects, what's waiting on others and what's next, with a few sentences on how it went;
  • where your health care stands: a timeline, what's still open, what to say on the phone, and a one-page brief for your next appointment;
  • the forms you're handed (a PDF application, an intake form) filled from your records, leaving the ID numbers, account numbers and signature to you;
  • advice grounded in your own patterns;
  • once a week, what's worth taking on (a call for papers, a fellowship, a talk, a piece of writing, with a first step drafted) and what's worth letting go of (a meeting, a thread, an invitation, with the decline drafted), by your own goals;
  • where each of your long-running projects stands, from your messages and its code on every branch, and, given its code, the next step of its work done on a branch of its own for you to accept;
  • a map of the people you know and how they connect;
  • a searchable memory that brings the right past messages and notes to mind, by their words and by their meaning;
  • drafts of replies, emails and plans, written in your voice: a reply for each chat waiting on you, in the way you write to that person, after reading any link they sent.

It drafts; by default it never sends, posts or buys anything by itself. You copy a draft or open it in its app with the text filled in, and press send yourself; once your reply shows up in that chat, the draft is marked sent. Chats that need no answer (thanks, a receipt, an auto-reply) get no draft. A draft you changed before sending, or rewrote on the page with "Write my own", is shown, beside what you sent, to the drafts written for you after it, and a draft your messages since made moot (someone else did it, or the plan changed) is folded away with why. What you tell it on the home page shapes every draft: a fact your messages don't show ("I'm not going on Tuesday"), kept for two weeks, or a rule kept until you remove it ("keep replies to my manager short"). Drafts never leave a blank for you to fill in. An outbox you switch on per app lets replies go out on their own, only where that app's drafts have gone out nearly unchanged many times, with a window to cancel each one, never to someone new, and never with a link, an address or a number the chat didn't already have (docs/memory.md). It places a phone call only when you press Call: it waits on hold, says it's an automated assistant calling for you, and rings you when a person answers (docs/calls.md). The longer-term plan, moving from drafting to acting on approval to acting within rules you set, is in docs/roadmap.md.

A Privacy view on the page shows what stays on your computer, every call to a model in the last 30 days and where it ran (on your computer or at Anthropic), and what's read from elsewhere. Other people's words go into prompts marked as data, never as instructions.

The name: in the Greek city-states, a proxenos was the citizen who formally acted on someone else's behalf.

What it looks like

These come from proxenos demo (see Quick start), like the home page at the top: a made-up person, Joe Bloggs, and his made-up records. None of it is real.

| Care: where things stand, what's yours to do, what to say | Errands: carried as far as they go, one tap short of sending | |---|---| | Care: a summary of a knee injury's referrals, scan and bill, and the next steps, each with a due date | Errands: a message to a recruiter with the CV attached, ready to send, with how the address was found | | Drafts: replies and documents in your voice, ready to send | Tasks: what was asked of you, pulled from your conversations | | Drafts: a WhatsApp reply and an email reply to send, with each app's track record and the outbox, off | Tasks: tasks from email and WhatsApp, with who asked, when, and when each is due |

Journal: the day, written down, from what Proxenos already tracks

Journal: a brief on yesterday, the next three things to do, and the replies sent, what's waiting on others and what's next

On a phone, over proxenos phone:

<p> <img src="docs/images/home-phone.png" width="15.5%" alt="Home on a phone"> <img src="docs/images/care-phone.png" width="15.5%" alt="Care on a phone"> <img src="docs/images/errands-phone.png" width="15.5%" alt="Errands on a phone"> <img src="docs/images/drafts-phone.png" width="15.5%" alt="Drafts on a phone"> <img src="docs/images/tasks-phone.png" width="15.5%" alt="Tasks on a phone"> <img src="docs/images/journal-phone.png" width="15.5%" alt="Journal on a phone"> </p>

How it works

your sources ──ingest──▶ vault/raw ──observe──▶ notes ──consolidate──▶ self-model
                              │                                             │
                              └──index──▶ memory (full-text search)         ▼
                                                        clones: feed, tasks, advice,
                                                        choices, watch, network,
                                                        recall, projects, project
                                                        steps, errands, drafts,
                                                        journal
                                                                │
                                                        local page (127.0.0.1)
  • Ingesters pull each source into the vault as monthly files. Re-running one is always safe.
  • Observers read one month of one source (only your own side of conversations) and write notes about you.
  • The master merges the notes into a single self-model: identity, interests, taste, voice, routines and open questions. Slow-changing parts change only through proposals you accept.
  • Clones read the self-model and each do one job. The page shows their output and records your reactions, which feed back into the next run.

More detail: architecture, memory, data sources, the feed and the page, a voice model trained on your own writing, related work.

Privacy

  • Code and data are separate. This repository holds code only. Everything personal lives in a private vault outside the repo (picked on the first-run setup page, or PROXENOS_VAULT; default ~/proxenos_vault). The vault keeps its own local git history of how the self-model changes.
  • What reaches the model: observers send the model only your side of conversations, never other people's messages. Clones send your self-model, plus public items such as papers and news. The rest send only the records their job needs, picked out in code:
    • tasks: conversations you're part of, with the title and opening of each draft waiting on you, and what you've told Proxenos on the home page;
    • care: only records about getting health care, with people in chats replaced by labels and phone, account and member numbers blanked;
    • drafts: the recent conversation in each direct chat, or Slack thread that mentions you or that you're in, waiting on your reply, with what your memory holds about that person and what the chat depends on (the newest records naming them, and the results of the searches the model asks for), what you've told Proxenos (your rules, and recent notes) and your calendar for the coming week (each event's time and title);
    • choices: the mail and Slack posts that announce an opening (a call, a fellowship, a talk), with who sent each, and what your time goes to: meetings, unanswered invitations, and the last message of each chat that has waited on you for over a week;
    • projects: the records that name each project or come from its people, and its code's commit messages and changed documents; a project's steps also give its code and data to the runner that takes the step,

Truncated for display — read the full file on GitHub.

Related Skills

View on GitHub
GitHub Stars3
CategoryDevelopment
Updated1h ago
Forks0

Languages

Python

Trust signals

92/100

From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.

1 low
proxenos — MCP Server: Install & Safety Check | SkillAgent