SkillAgentSearch skills...

synthetick-oss

Turn an investment thesis into audited, evidence-backed picks. MIT.

Install / Use

claude mcp add Matteoikarieth96 -- npx -y github:Matteoikarieth96/synthetick-oss

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

70/100

Supported Platforms

Claude Code
Claude Desktop
Zed

Our assessment of synthetick-oss

synthetick-oss scores 70/100 on our quality scale, 873rd of 945 AI & Machine Learning skills we index.

Its MCP Server is 5.8 KB long, split into 6 sections with 1 code example: a solid amount of guidance for an agent.

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

Substance
26/30
Structure
15/20
Description
12/15
Adoption
3/20
Freshness
15/15

Maintenance, license and trust

  • The repository was last updated today, so synthetick-oss 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 87/100, with 2 cautions 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.

Safety scan

No issues found

Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful.

AI review by kimi-k2.7-code on 2026-11-01. Automated pattern scan on 2026-11-01. It catches known dangerous patterns, not every risk — read a skill before letting an agent act on it.

synthetick-oss compared with similar skills

All 4 of these similar skills score higher than synthetick-oss; compare them before choosing.

SkillScoreStarsUpdatedFormat
synthetick-oss (this skill)by Matteoikarieth96703todayMCP Server
claude-memby thedotmack100108.0ktodayCLAUDE.md
Agent-Reachby Panniantong100107.3k7d agoCLAUDE.md
Understand-Anythingby Egonex-AI10089.4k6d agoCLAUDE.md
headroomby headroomlabs-ai10078.5k1d agoCLAUDE.md

Frequently asked questions

How do I install synthetick-oss?
Run claude mcp add Matteoikarieth96 -- npx -y github:Matteoikarieth96/synthetick-oss. The install tabs above show the steps for each supported agent.
Which AI agents does synthetick-oss work with?
It is written for Claude Code, Claude Desktop and Zed, as a MCP Server file. Other agents that read the same format can often use it too.
Is synthetick-oss safe to use?
Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful. It is MIT-licensed and scores 87/100 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 synthetick-oss still maintained?
The repository was last updated today, so synthetick-oss is actively maintained.
<h1 align="center">SyntheTick</h1> <p align="center"><b>Turn an investment thesis into audited, evidence-backed picks, without invented numbers.</b><br> Open source, MIT licensed. Live at <a href="https://synthetick.org">synthetick.org</a>.</p> <p align="center"> <a href="LICENSE"><img alt="License: MIT" src="https://img.shields.io/github/license/Matteoikarieth96/synthetick-oss"></a> <a href="https://github.com/Matteoikarieth96/synthetick-oss/actions/workflows/ci.yml"><img alt="CI" src="https://github.com/Matteoikarieth96/synthetick-oss/actions/workflows/ci.yml/badge.svg?branch=main"></a> <a href="https://github.com/Matteoikarieth96/synthetick-oss/releases"><img alt="Latest release" src="https://img.shields.io/github/v/release/Matteoikarieth96/synthetick-oss"></a> <a href="https://github.com/Matteoikarieth96/synthetick-oss/issues?q=is%3Aissue+is%3Aopen+label%3A%22good+first+issue%22"><img alt="Good first issues" src="https://img.shields.io/github/issues/Matteoikarieth96/synthetick-oss/good%20first%20issue?label=good%20first%20issues"></a> </p>

SyntheTick in 80 seconds: idea in, audited assets out

Watch the full video (MP4). It is silent, with captions, and was made with whiteboard-video, an open-source Claude Code skill that turns any topic into a hand-drawn explainer.

You paste a thesis, an article, a tweet or a PDF. SyntheTick pulls out the idea and the rules you set ("only crypto", "European ETFs only", "no defense"), searches a universe of stocks, ETFs, bond funds, crypto and selected pre-IPO companies, ranks up to ten assets with a 0 to 100 alignment score and a reason, audits every pick twice, and attaches real market data where the data licence allows it. It is a research starting point, not investment advice.

The honesty contract

  • A requirement from your thesis is enforced three times: a SQL hard filter, a deterministic code audit, and an LLM audit against your own words.
  • If nothing qualifies, the answer is "Nothing matched your requirements". Requirements are never relaxed quietly.
  • Every number carries an "as of" time. If a vendor has no data, the card shows a dash. Nothing is simulated.

How it works

<p align="center"><img src="docs/assets/architecture.svg" alt="Architecture: data sources, ingest, Postgres with pgvector, a six step runtime pipeline, and five surfaces" width="900"></p>

| Folder | What lives there | |---|---| | ingest/ | Scheduled data jobs: normalise, dedupe by ISIN, quality gates, embeddings | | runtime/ | The pipeline: thesis extraction, retrieval, rerank, two audits, analysis, market data. Prompts live here as plain strings | | server/ | HTTP server: web API with live status stream, public /v1 API, MCP server, rate limits | | app/, public/ | The web app (Next.js shell plus the interaction layer) | | bot/ | The X bot | | db/ | Postgres schema, RLS, functions. See db/README.md | | sail-agent/ | Experimental autonomous agent that trades tokenized stocks. Can move real funds. Separate from the app, read its README first | | docs/ | Documentation, API and MCP, quickstart, test reports |

The single source of truth for behaviour is SPEC.md. When code and spec disagree, the spec wins; if you change a decision, change the spec first.

Run it

The cheapest honest setup is crypto only: a free Supabase project, a Voyage key, an OpenRouter key and a few cents of LLM credit per run. Follow the quickstart.

npm install
npm run typecheck && npm run test:offline   # no keys, no database, no .env needed
cp .env.example .env     # then fill in the REQUIRED values to run the app

Use it from code

A public HTTP API and an MCP server for AI agents (same API keys, 1 credit per screen). See docs/API.md.

Contribute

<p align="center"><img src="docs/assets/contribute.svg" alt="Contribute in six steps, and where help is wanted" width="900"></p>

Read CONTRIBUTING.md. Pull requests target staging. Look for the good first issue label. Security problems go through the private report on the Security tab, see SECURITY.md. Running it for other people? Read docs/PRIVACY.md. Be kind: CODE_OF_CONDUCT.md.

Data, licenses and trademarks

  • No vendor data payloads are in this repository. The only data file is a public on-chain registry (runtime/universe/robinhood-chain.json: token contracts from Robinhood's public stock-token API, decimals and holder counts read from the chain explorer). Bring your own keys and read each vendor's terms before you display or redistribute their data. Equity and ETF data vendors usually require a separate display license for public apps, so the app hides that data unless you say you hold one (see .env.example).
  • Attribution is required for some free sources. If you show CoinGecko API data, show "Data provided by CoinGecko" with a link to https://www.coingecko.com/en/api next to it; for GeckoTerminal data, "On-chain data provided by GeckoTerminal" with a link to https://www.geckoterminal.com. The app does this out of the box.
  • The code is MIT, except the brand assets in brand/. Third-party files keep their own notices: see NOTICE.md and THIRD_PARTY.md.
  • The SyntheTick name, logo and Dino mascot are not licensed for use as your own brand, see TRADEMARKS.md. Other company names are used only to say which service a piece of code talks to; no affiliation or endorsement is implied.
  • Nothing here is investment, legal or tax advice. Tokenized securities and crypto assets can be restricted where you live, and sail-agent/ can move real funds.

Related Skills

View on GitHub
GitHub Stars3
CategoryAI
Updated13h ago
Forks3

Languages

TypeScript

Trust signals

87/100

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

2 low