Skip to content
Back to skills

Three Ws Discovery

ASecurity

Read live three.ws discovery data (trending agents, trending coins, the $THREE holder leaderboard and tiers, the activity ticker) through the activity-mcp tools. Use when asked what is hot on three.ws, where a wallet ranks, or what just happened on the platform.

  • 274 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added November 11, 2026
ai-agentsrustapi

Works with

  • api
  • mcp

Security analysis

A100/100

Scanned November 11, 2026

npx -y skills add nirholas/three.ws --skill three-ws-discovery --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Three Ws Discovery?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Three Ws Discovery
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/nirholas-three-ws-discovery-three-ws/badge)](https://www.skillsdirectory.com/skills/nirholas-three-ws-discovery-three-ws)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: three-ws-discovery
description: Read live three.ws discovery data (trending agents, trending coins, the $THREE holder leaderboard and tiers, the activity ticker) through the activity-mcp tools. Use when asked what is hot on three.ws, where a wallet ranks, or what just happened on the platform.
---

# three.ws live discovery

The plugin starts `@three-ws/activity-mcp` over stdio (`npx -y @three-ws/activity-mcp`). It is a read-only wrapper over the public three.ws API. It needs no key, signer or payment.

## Environment (both optional)

- `THREE_WS_BASE`: API origin, default `https://three.ws`. Change it only to target a self-hosted or preview deployment.
- `THREE_WS_TIMEOUT_MS`: per-request timeout in milliseconds, default `20000`. Must be a positive number.

## Tools

| Tool | Arguments | Backing endpoint |
| --- | --- | --- |
| `get_trending_agents` | `window` (`24h`, `7d`, `all`; default `24h`), `sort` (`activity` or `trust`), `limit` (1-20, default 10) | `GET /api/trending` |
| `get_trending_coins` | `limit` (1-20, default 10) | `GET /api/trending` |
| `get_holder_leaderboard` | `limit` (1-100, default 50), `offset` (default 0), `wallet` (base58, optional) | `GET /api/leaderboard` |
| `get_tier_info` | none | `GET /api/leaderboard` (tier ladder) |
| `get_feed_events` | `limit` (1-100, default 30) | `GET /api/feed` |

Every tool returns JSON text with `ok: true` on success. A failure returns `ok: false`, an `error` code and a `message`, with the result flagged as an error.

## How to use them

1. For "what is trending", call `get_trending_agents` (ranked by real chat activity) or `get_trending_coins` (ranked by the Oracle conviction score, 0-100). Coin results are discovery data, not financial advice. Say so when you summarize them.
2. For "where does this wallet rank", call `get_holder_leaderboard` with `wallet` set. The `you` field reports that wallet's rank, amount, share of supply and tier even when it is outside the returned page. Page through the board with `limit` and `offset`.
3. For "what does tier X mean" or "how far to the next tier", call `get_tier_info`. Each tier is `{ id, label, min, accent }`, where `min` is the whole-token $THREE balance required.
4. For recent platform events, call `get_feed_events`. Events carry `type`, `ts` (epoch milliseconds), an already-public `actor` label and type-specific fields.

## Worked example

Request: "Which agents are hottest this week, weighted toward trusted wallets?"

Call `get_trending_agents` with `{ "window": "7d", "sort": "trust", "limit": 5 }`. Report each row's `name`, `window_chats`, `chat_count` and `agent_url`, and the reputation card `{ score, tier }` when present.

## Notes

- Data moves between calls. Do not cache rankings across a conversation.
- Nothing here signs or spends. If the user wants to act on a result, that is a separate tool and needs their explicit confirmation.

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

View all in ai-agents →

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.