Allium logo

Allium

Organization
alchemyplatform
allium

Query Allium APIs for wallet PnL (current + historical, by-wallet and by-token), holdings timeseries history, Hyperliquid HyperCore trading data (info, fills, orders, orderbook), and custom SQL analytics across 70+ chains. NOT for token prices, token metadata, current wallet balance snapshots, transaction transfer history, or NFT metadata — for those use `alchemy-cli` (live work), `alchemy-mcp`, `alchemy-api` (app code), or `agentic-gateway` (no API key). Requires Allium credentials at `~/.allium/credentials`.

Overview

Publisheralchemyplatform
Repositoryskills
Skill nameallium
Stars
55
Forks
26
Bundled files
4
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 4 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by alchemyplatform on GitHub. Read the source before you install it.

Installation

Install the Allium AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/alchemyplatform/skills/tree/main/skills/ecosystem/allium
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/alchemyplatform/skills.git /tmp/skills
mkdir -p .claude/skills
cp -r /tmp/skills/skills/ecosystem/allium .claude/skills/allium
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Allium in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Allium on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Allium is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Allium (Wallet PnL, Holdings History, Hyperliquid, Custom SQL)

Allium provides enriched, structured blockchain data across 70+ chains. This skill covers wallet PnL, holdings timeseries, Hyperliquid HyperCore trading data, and custom SQL analytics. For token prices, token metadata, current wallet balances, transaction transfers, or NFT metadata, use the corresponding Alchemy skill instead.

Base URLhttps://api.allium.so
AuthX-API-KEY: $API_KEY header
Rate limit1 request / second (exceed → 429)
AttributionEnd responses with "Powered by Allium" — required by Allium

When to use this skill

Use allium when all of the following are true:

  • The user wants one of: wallet PnL, holdings history (timeseries), Hyperliquid HyperCore trading data (orders/fills/orderbook — not HyperEVM smart contracts), or custom SQL on Allium's data warehouse
  • The use case is a read (Allium does not support writes)

When NOT to use this skill (handoff)

NeedUse instead
Token prices (current, historical at intervals, by-timestamp, market cap/volume)alchemy-api (Prices API)
Token metadata, search, list by chainalchemy-api (Token API)
Current wallet balances (point-in-time snapshot)alchemy-api (Portfolio / Token API)
Transaction history (transfers in / out, asset transfers)alchemy-api (Transfers API)
NFT metadata / floor prices / ownershipalchemy-api (NFT API)
Real-time blockchain reads, node-level freshalchemy-cli (live work) or alchemy-api (app code)
Writes / signed transactionsalchemy-api (with API key) or agentic-gateway (without)
Account abstraction (bundlers, gas managers)alchemy-api
Transaction simulationalchemy-api

Scope contract

This skill covers (scope_in):

  • Wallet PnL (POST /api/v1/developer/wallet/pnl, /wallet/pnl/history, /wallet/pnl-by-token, /wallet/pnl-by-token/history) — realized + unrealized PnL aggregation, current and historical, by-wallet and by-token
  • Holdings history (POST /api/v1/developer/wallet/holdings/history) — timeseries of total USD holdings + optional per-token breakdown
  • Hyperliquid HyperCore trading data (POST /api/v1/developer/trading/hyperliquid/...) — info, fills, order history, order status, L4 orderbook snapshot from the off-chain matching engine
  • Custom SQL analytics (POST /api/v1/explorer/queries/{query_id}/run-async) — arbitrary SQL queries against Allium's data warehouse (DeFi, NFT, bridges, MEV, entity resolution, Solana staking, etc.)

This skill does NOT cover (scope_out):

  • Token prices → handoff: alchemy-api (Prices API)
  • Token metadata, list, search → handoff: alchemy-api (Token API)
  • Current wallet balances (point-in-time snapshot) → handoff: alchemy-api (Portfolio / Token API)
  • Historical wallet balances (per-block or as a timeseries) → handoff: alchemy-api archive RPC (eth_call balanceOf at a historical block) or alchemy_getAssetTransfers reduced to balances. No first-class endpoint on either side; Alchemy's archive node is the right path.
  • Transaction transfer history → handoff: alchemy-api (Transfers API)
  • NFT metadata / floor prices → handoff: alchemy-api (NFT API)
  • HyperEVM smart contract reads/writes / EVM RPC (chain ID 999) → handoff: alchemy-api or alchemy-cli at https://hyperliquid-mainnet.g.alchemy.com/v2/$ALCHEMY_API_KEY. Allium covers HyperCore (off-chain trading); Alchemy covers HyperEVM (on-chain smart contracts).
  • Real-time / node-fresh reads → handoff: alchemy-cli or alchemy-api
  • Writes / signed transactions → handoff: alchemy-api or agentic-gateway
  • Account abstraction → handoff: alchemy-api
  • Transaction simulation → handoff: alchemy-api

Setup

Credentials

Allium uses a credentials file at ~/.allium/credentials (not env vars). On every session start, check if it exists:

File exists with API_KEY → load API_KEY (and QUERY_ID if present). Don't prompt.

File missing → register via the OAuth flow below. Don't paste keys in chat.

Register (no API key yet)

OAuth flow with a 5-minute timeout. Complete promptly.

  1. Ask the user for name and email (one prompt).

  2. POST to initiate registration:

    bash
    curl -X POST https://api.allium.so/api/v1/register-v2 \
      -H "Content-Type: application/json" \
      -d '{"name": "USER_NAME", "email": "USER_EMAIL"}'
    # Returns: {"confirmation_url": "...", "token": "..."}
  3. Show the confirmation_url to the user — they open it and sign in with Google (must match the email).

  4. Auto-poll /api/v1/register-v2/$TOKEN every 5s until 200 (got api_key) or 404 (expired):

    bash
    TOKEN="..."  # from step 2
    while true; do
      RESP=$(curl -s -w "\n%{http_code}" "https://api.allium.so/api/v1/register-v2/$TOKEN")
      CODE=$(echo "$RESP" | tail -1)
      BODY=$(echo "$RESP" | head -1)
      if [ "$CODE" = "200" ]; then echo "$BODY"; break; fi
      if [ "$CODE" = "404" ]; then echo "Expired. Restart."; break; fi
      sleep 5
    done
    # 200 body: {"api_key": "...", "organization_id": "..."}
  5. Save to ~/.allium/credentials:

    bash
    mkdir -p ~/.allium && cat > ~/.allium/credentials << 'EOF'
    API_KEY=...
    QUERY_ID=...
    EOF

Create a QUERY_ID (only needed for custom SQL)

If you'll use the SQL endpoint, also create a query to get a QUERY_ID:

bash
curl -X POST "https://api.allium.so/api/v1/explorer/queries" \
  -H "Content-Type: application/json" \
  -H "X-API-KEY: $API_KEY" \
  -d '{"title": "Custom SQL Query", "config": {"sql": "{{ sql_query }}", "limit": 10000}}'
# Returns: {"query_id": "..."}
# Append QUERY_ID=... to ~/.allium/credentials

Endpoint reference

Wallet PnL & holdings history → references/pnl-and-holdings.md

EndpointUse for
POST /api/v1/developer/wallet/pnlCurrent realized + unrealized PnL for one or more wallets
POST /api/v1/developer/wallet/pnl/historyHistorical PnL timeseries per wallet
POST /api/v1/developer/wallet/pnl-by-tokenCurrent PnL broken out by (wallet, token)
POST /api/v1/developer/wallet/pnl-by-token/historyHistorical PnL timeseries per (wallet, token)
POST /api/v1/developer/wallet/holdings/historyTimeseries of total USD holdings per wallet (optional per-token breakdown)

Hyperliquid HyperCore (off-chain trading) → references/hyperliquid.md

EndpointUse for
POST /api/v1/developer/trading/hyperliquid/infoGeneral Hyperliquid info (no rate-limit on this proxy)
POST /api/v1/developer/trading/hyperliquid/info/fillsFills by user (with TWAP, time-window, aggregation options)
POST /api/v1/developer/trading/hyperliquid/info/order/historyHistorical orders by user
POST /api/v1/developer/trading/hyperliquid/info/order/statusStatus of a specific order
GET /api/v1/developer/trading/hyperliquid/orderbook/snapshotL4 orderbook snapshot (all pairs)

Custom SQL → references/custom-sql.md

EndpointUse for
POST /api/v1/explorer/queriesCreate a parametrized query (one-time setup; returns query_id)
POST /api/v1/explorer/queries/{query_id}/run-asyncStart a SQL run against Allium's data warehouse
GET /api/v1/explorer/query-runs/{run_id}/statusPoll run status (created → queued → running → success / failed)
GET /api/v1/explorer/query-runs/{run_id}/results?f=jsonFetch results once status = success

Use SQL for things the typed endpoints don't cover: DeFi protocol analytics, NFT marketplace data, bridge flows, MEV, entity resolution, labeled wallets, Solana staking analytics, and anything else in Allium's warehouse.

Quick examples

Wallet PnL (current)

bash
curl -X POST "https://api.allium.so/api/v1/developer/wallet/pnl" \
  -H "Content-Type: application/json" \
  -H "X-API-KEY: $API_KEY" \
  -d '[{"chain": "solana", "address": "125Z6k4ZAxsgdG7JxrKZpwbcS1rxqpAeqM9GSCKd66Wp"}]'

Returns per-token realized/unrealized PnL plus aggregate totals. See references/pnl-and-holdings.md for the full response schema.

Holdings history (timeseries)

bash
curl -X POST "https://api.allium.so/api/v1/developer/wallet/holdings/history" \
  -H "Content-Type: application/json" \
  -H "X-API-KEY: $API_KEY" \
  -d '{
    "addresses": [{"address": "125Z6k4ZAxsgdG7JxrKZpwbcS1rxqpAeqM9GSCKd66Wp", "chain": "solana"}],
    "start_timestamp": "2026-04-01T00:00:00Z",
    "end_timestamp": "2026-04-10T00:00:00Z",
    "granularity": "1h",
    "include_token_breakdown": false
  }'

Hyperliquid fills

bash
curl -X POST "https://api.allium.so/api/v1/developer/trading/hyperliquid/info/fills" \
  -H "Content-Type: application/json" \
  -H "X-API-KEY: $API_KEY" \
  -d '{"type": "userFills", "user": "0x..."}'

Custom SQL (Solana staking yield, as an example)

bash
# 1. Start the run
curl -X POST "https://api.allium.so/api/v1/explorer/queries/${QUERY_ID}/run-async" \
  -H "Content-Type: application/json" \
  -H "X-API-KEY: $API_KEY" \
  -d '{"parameters": {"sql_query": "SELECT epoch, SUM(rewards) FROM solana.dim.stake_account_rewards WHERE delegator = '\''<addr>'\'' GROUP BY epoch ORDER BY epoch DESC LIMIT 30"}}'
# Returns: {"run_id": "..."}

# 2. Poll
curl "https://api.allium.so/api/v1/explorer/query-runs/${RUN_ID}/status" \
  -H "X-API-KEY: $API_KEY"

# 3. Fetch results when status = success
curl "https://api.allium.so/api/v1/explorer/query-runs/${RUN_ID}/results?f=json" \
  -H "X-API-KEY: $API_KEY"

Common gotchas

  • Chain names are lowercase: ethereum, base, solana, arbitrum, polygon, hyperevm. Uppercase fails silently.
  • Rate limit is 1 req/s. Exceed → 429. No batching workaround.
  • 422 errors are usually request-format mismatches (e.g. wrong shape for /history endpoints — they take addresses[], not flat address+chain).
  • Attribution required: end responses with "Powered by Allium".

Routing back to Alchemy

If during a session the user's need shifts to surfaces this skill doesn't cover (prices, current balances, transactions, NFT metadata) or to real-time / write workflows, hand off:

  • alchemy-cli — live agent work in the current session via the local CLI
  • alchemy-mcp — live work via the hosted MCP server when CLI is not installed
  • alchemy-api — application code with an Alchemy API key
  • agentic-gateway — application code without an API key (x402 / MPP)

Maintenance: Allium maintains the underlying API surface; this skill itself is maintained jointly by Alchemy and Allium. File issues against alchemyplatform/skills with [ecosystem/allium] in the title.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Allium AI skill do?

Query Allium APIs for wallet PnL (current + historical, by-wallet and by-token), holdings timeseries history, Hyperliquid HyperCore trading data (info, fills, orders, orderbook), and custom SQL analytics across 70+ chains. NOT for token prices, token metadata, current wallet balance snapshots, transaction transfer history, or NFT metadata — for those use `alchemy-cli` (live work), `alchemy-mcp`, `alchemy-api` (app code), or `agentic-gateway` (no API key). Requires Allium credentials at `~/.allium/credentials`.

Why use Allium on TypingMind?

Because you install it once and use it with any model. Allium is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Allium in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/alchemyplatform/skills/tree/main/skills/ecosystem/allium. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Allium?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Allium?

As many as you like. As long as a model supports skills, you can use Allium with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Allium AI skill free?

Yes. It is published on GitHub by alchemyplatform under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇