frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

Show HN: DollhouseMCP 2.0, open-source MCP composable AI building blocks

https://dollhousemcp.com/
3•mickdarling•3h ago
Hi, I'm Mick. I've been building this for the last 9 months

DollhouseMCP 2.0 is an open-source MCP server for making and using composable building blocks for AI customization. You build elements as portable MD and YAML files, compose them into stacks, and activate the stacks in any MCP-compatible client.

Element types that drive behavior and permissions:

* Personas: behavioral profiles (how the AI sounds and acts)

* Skills: discrete capabilities (what the AI can do)

* Agents: goal-oriented, multi-step executors

* Ensembles: composed stacks of the above

* Plus templates for structured outputs and memories for persistent context.

Two things I think are actually new here:

1. Identity-based permissioning. When you activate a persona, skill, agent, or ensemble, its permission policy takes effect in the server. Same client, same LLM, different permission surface depending on which active elements are loaded. A read-only analyst persona blocks creates and deletes regardless of what the client allows. A security-focused ensemble can deny specific destructive operations. This runs server-side, after the client approves the call, so policy cannot be overridden by the LLM or the client.

2. A bimodal agent loop. Agents do not run free inside the LLM. Every step hands control back to the MCP server, which evaluates the proposed operation against the active permission stack, runs autonomy and risk checks, enforces any hard blocks, then returns a decision to the LLM with continue, pause, or escalate guidance. The LLM acts on what it is allowed to do, describes the next step, and hands back to the server. The loop repeats until the goal completes or a human is asked to intervene. Higher agency stays observable and bounded instead of opaque.

There's also a audit trail of approved and denied actions a danger zone lockout that will prevent the LLM from doing truly dangerous things if they go through the MCP server. And any active Dollhouse agent that is running through the DollhouseMCP server has it's actions evaluated every step.

I added easy configuration through the web console for a wide variety of MCP clients if you use the one-liner, and there are logs and metrics as well as the local and github hosted portfolio and collection to save your Dollhouse elements and share and use other's. They are all validated and scanned several times along the their distribution path to keep things as safe as we can.

The one-liner install: npx @dollhousemcp/mcp-server@latest --web

Happy to go deep on the permissioning model, the bimodal agent loop, composition patterns, YAML schema, or anything else.

Home: https://dollhousemcp.com Repo: https://github.com/DollhouseMCP/mcp-server Collection: https://collection.dollhousemcp.com

Hope you like it.

Routines in Claude Code

https://claude.com/blog/introducing-routines-in-claude-code
1•meetpateltech•1m ago•0 comments

RAG-Forge – Framework-agnostic RAG toolkit with a maturity model

https://github.com/hallengray/rag-forge
1•hallengray•1m ago•0 comments

The Perimeter Problem

https://www.southwind.ai/blog/the-perimeter-problem
1•Eswo•1m ago•0 comments

Show HN: Java RocksDB Without JNI

https://github.com/dfa1/rocksdbffm
1•dfa11•1m ago•0 comments

I Discover New Blogs

https://kevquirk.com/how-i-discover-new-blogs
2•speckx•2m ago•0 comments

William Cecil's Succession Plan

https://www.historytoday.com/archive/history-matters/william-cecils-succession-plan
1•Petiver•2m ago•0 comments

Can LLMs Perform Synthesis?

https://arxiv.org/abs/2603.20264
3•PaulHoule•4m ago•0 comments

Add Animal Crossing events to your digital calendar

https://sethmlarson.dev/animal-crossing-calendar
2•SethMLarson•4m ago•0 comments

Show HN: Privacy-first retirement planner – no bank linking, AI-powered analysis

https://planwithclarity.app
2•smilano85•4m ago•0 comments

The problem with thinking you're part Neanderthal

https://www.technologyreview.com/2026/04/14/1135169/problem-thinking-part-neanderthal-human-evolu...
2•laurex•5m ago•1 comments

Anthropic Opposes the Extreme AI Liability Bill That OpenAI Backed

https://www.wired.com/story/anthropic-opposes-the-extreme-ai-liability-bill-that-openai-backed/
3•srameshc•5m ago•0 comments

Tell HN: GitHub might have been leaking your webhook secrets. Check your emails.

3•ssiddharth•5m ago•0 comments

Interview: John Calhoun on the Origins of Glider (2010)

https://macscene.net/d/4678-interview-john-calhoun-on-the-origins-of-glider-part-1
2•CharlesW•6m ago•0 comments

Anthropic Plots Lovable Challenger

https://sifted.eu/articles/anthropic-lovable-challenger-leak
2•AnhTho_FR•7m ago•0 comments

Digital Realty Spending €2B to Turn Italy into a Mediterranean Data Hub

https://thecoolingreport.com/intel/digital-realty-italy-2-billion-mediterranean-data-hub.html
2•jackdilusso•7m ago•1 comments

The Cost of Concurrency Coordination [video]

https://www.youtube.com/watch?v=tND-wBBZ8RY
2•tosh•9m ago•0 comments

A+

https://aplwiki.com/wiki/A%2B
2•tosh•10m ago•0 comments

PicoTop – A basic computer and OS based on the RP2350 microcontroller

https://github.com/7west/PicoTop
2•rickcarlino•12m ago•0 comments

The Dead Internet Theory Is Real. Now What?

https://twitter.com/abrilzucchi/status/2011850529140564278
3•abruzucchi•12m ago•0 comments

OpenAI has bought AI personal finance startup Hiro

https://techcrunch.com/2026/04/13/openai-has-bought-ai-personal-finance-startup-hiro/
4•Brajeshwar•13m ago•1 comments

Show HN: AfterSession – AI tool to help therapists write notes faster

https://aftersession.ai/
2•edwardsrobbie•13m ago•1 comments

Show HN: We built an AI Agent to reproduce bugs

https://www.metabase.com/blog/reprobot-github-issue-triage-agent
4•nvoxland•13m ago•0 comments

Bee write back Build Guide [video]

https://www.youtube.com/watch?v=JutsTp7yeNU
2•ofrzeta•13m ago•0 comments

Google develops its own desktop Agent to compete with Cowork

https://www.testingcatalog.com/google-develops-its-own-desktop-agent-to-compete-with-cowork/
3•gmays•13m ago•0 comments

I'm a PM Who Built an AI Workflow Without Writing Code (Here's What I Learned)

https://www.arthur.ai/blog/what-building-an-agent-actually-means-and-why-most-people-get-it-wrong
4•ashleynader•13m ago•0 comments

Cheapest way to keep a UK mobile number using an eSIM

https://shkspr.mobi/blog/2026/04/cheapest-way-to-keep-a-uk-mobile-number-using-an-esim/
2•worldofmatthew•14m ago•0 comments

Palm Biometrics for Verifying AI Agents

https://very.org/blog/what-is-very-ai-palm-biometrics-deepfake-detection/
3•abruzucchi•14m ago•1 comments

Show HN: Plotline, an open-source initiative to help you own your health data

https://plotline.care/
2•kalturnbull•15m ago•0 comments

Show HN: A tool to compare LLM costs for typical coding domains

https://the-designengineer.com/http://localhost:4321/model-cost-estimator/
2•canxerian•15m ago•0 comments

Booking.com warns customers of hack that exposed their data

https://www.theguardian.com/technology/2026/apr/13/booking-com-customers-hack-exposed-data
3•luispa•16m ago•0 comments