clawrouter · git:20261028.44c5331 · 2026-10-28 · sha256 c4501f65e0fbb7d6
clawrouter git:20261028.44c5331A
Immutable. This exact content never changes and is served at /api/v1/blob/c4501f65e0fbb7d6.
---
name: clawrouter
description: Hosted-gateway LLM router — save 84% on inference costs. A local proxy that forwards each request to the blockrun.ai gateway, which routes to the cheapest capable model across 90 models from OpenAI, Anthropic, Google, DeepSeek, xAI, Z.AI, and more. Free open-weight models included. Also exposes realtime market data (global stocks, crypto, FX, commodities), Twitter/X intelligence, prediction-market data across Polymarket, Kalshi, Limitless, Opinion, Predict.Fun, dFlow + UMA oracle resolution + wallet identity & clustering, phone-number intelligence (carrier + SIM-swap fraud detection) plus AI-powered outbound voice calls (Twilio + Bland.ai) as built-in agent tools. Not a local-inference tool — prompts are sent to the blockrun.ai gateway.
triggers:
- "clawrouter"
- "claw router"
- "@blockrun/clawrouter"
- "blockrun gateway"
- "blockrun llm router"
- "blockrun ai gateway"
- "blockrun.ai inference"
- "save on llm costs blockrun"
- "cheapest model blockrun"
- "free nvidia models blockrun"
- "x402 llm payment"
- "usdc llm gateway"
- "openrouter alternative"
- "blockrun phone"
- "blockrun voice call"
- "ai phone call"
- "phone number verification"
- "sim swap detection"
- "outbound ai call"
homepage: https://blockrun.ai/clawrouter.md
repository: https://github.com/BlockRunAI/ClawRouter
license: MIT
metadata:
{
"openclaw":
{
"emoji": "🦀",
"requires": { "config": ["models.providers.blockrun"] },
"install":
[
{
"id": "node",
"kind": "node",
"package": "@blockrun/clawrouter",
"bins": ["clawrouter"],
"label": "Install ClawRouter (npm)",
},
],
},
}
---
# ClawRouter
Hosted-gateway LLM router that saves <!-- br:savings.autoVsBaselinePct -->84<!-- /br:savings.autoVsBaselinePct -->% on inference costs by forwarding each request to the blockrun.ai gateway, which picks the cheapest model capable of handling it across <!-- br:models.chatVisible -->90<!-- /br:models.chatVisible --> models from 9 providers (<!-- br:models.free -->7<!-- /br:models.free --> free open-weight models). All billing flows through one BlockRun credential — either a USDC wallet paying x402 per call, or a BlockRun API key drawing on card-funded account credit. Either way you do not hold provider API keys.
**This is not a local-inference tool.** ClawRouter is a thin local proxy. Your prompts are sent over HTTPS to the blockrun.ai gateway for model execution. If your workload requires inference that never leaves your machine, use a local runtime like Ollama — ClawRouter is not the right tool for that use case.
Source: https://github.com/BlockRunAI/ClawRouter · npm: https://www.npmjs.com/package/@blockrun/clawrouter · License: MIT.
## Data Flow
```
Your app → localhost proxy (ClawRouter) → https://blockrun.ai/api (or sol.blockrun.ai/api, arc.blockrun.ai/api)
↓
OpenAI / Anthropic / Google / etc.
↓
Response → back through proxy → your app
```
**Sent to blockrun.ai on every request:** the model name, the full prompt/messages body, sampling params (temperature, max_tokens, tools, etc.), and a payment credential — an `X-PAYMENT` header containing a signed x402 USDC micropayment in wallet mode, or an `Authorization: Bearer brk_live_…` header in API-key mode.
**Not sent:** your wallet private key (only the detached payment signature is sent), any other local files, environment variables, or OpenClaw config beyond what's needed for this request.
**Blockrun's privacy stance:** https://blockrun.ai/privacy. Treat prompts the same way you'd treat prompts sent to any hosted LLM API (OpenAI, Anthropic, etc.) — do not send data you would not share with a third-party API provider.
## Credentials & Local Key Storage
ClawRouter does **not** collect or forward third-party provider API keys. You do not supply OpenAI, Anthropic, Google, DeepSeek, xAI, or NVIDIA credentials — the blockrun.ai gateway owns those relationships.
**What `models.providers.blockrun` stores (fully enumerated):**
| Field | Sensitive | Purpose |
| ----------- | ------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `walletKey` | Only in wallet mode | EVM private key used to sign USDC micropayments via x402. **Auto-generated locally on first run** — no user input required. Never transmitted over the network; only detached payment signatures are sent. Not used, read, or generated at all when a BlockRun API key is configured. |
| `solanaKey` | Yes | Solana keypair (BIP-44 `m/44'/501'/0'/0'`). Auto-derived from the same local mnemonic via `@scure/bip32` + `@scure/bip39`. |
| `gateway` | No | Gateway URL. Defaults: `https://sol.blockrun.ai/api` (Solana, default chain for new installs) · `https://blockrun.ai/api` (Base, default for pre-existing installs). |
| `routing` | No | Optional override of the router-core config (tier chains, `strategy: "rules"` rollback, `shadow` comparison, scorer keywords). |
**How and where keys are stored:**
- Keys live in the OpenClaw user config file — typically `~/.config/openclaw/config.json` on Linux, `~/Library/Application Support/openclaw/config.json` on macOS, `%APPDATA%\openclaw\config.json` on Windows — under the `models.providers.blockrun` path.
- Written by OpenClaw's standard config writer with `0600` permissions on POSIX systems (owner read/write only).
- **Stored in plaintext**, the same way every OpenClaw provider's API key is stored. ClawRouter does not add an extra encryption layer; your filesystem permissions are the security boundary. If you require an encrypted keystore, run OpenClaw on an encrypted volume (FileVault, LUKS, BitLocker) or use a dedicated burner wallet funded only with what you intend to spend.
- Auto-generation uses `@scure/bip39` to produce a 24-word mnemonic, then BIP-44 derivation for both chains. Source: [`src/wallet.ts`](https://github.com/BlockRunAI/ClawRouter/blob/main/src/wallet.ts).
**Operational guidance:** treat the wallet as a spending account with a small top-up, not a long-term store of value. Fund it with what you expect to spend on LLM calls. If the host machine is compromised, the wallet key is compromised — rotate and refund.
## Supply-Chain Integrity
- Every release is tagged on GitHub: https://github.com/BlockRunAI/ClawRouter/releases
- Every release publishes to npm with a matching version: https://www.npmjs.com/package/@blockrun/clawrouter?activeTab=versions
- The `skills/release/SKILL.md` mandatory checklist enforces: same version in `package.json`, matching git tag, matching GitHub release, and matching npm publish.
- To verify locally: `npm pack @blockrun/clawrouter@<version>` and compare the tarball contents to the tagged commit.
## Install
```bash
openclaw plugins install @blockrun/clawrouter
```
The structured `install` block above tells OpenClaw to install the auditable npm package `@blockrun/clawrouter`. Source for every version is on GitHub; every release is tagged.
## Setup
```bash
# Enable smart routing (auto-picks cheapest model per request)
openclaw models set blockrun/auto
# Or pin a specific model
openclaw models set openai/gpt-4o
```
## How Routing Works
Routing is done locally by [`@blockrun/router-core`](https://github.com/BlockRunAI/router-core) (Router Core V3.4, constraint-first), inlined into ClawRouter and pinned by commit. No network call, no second model — about a quarter of a millisecond per request.
1. A 15-dimension weighted scorer classifies each request into one of four tiers, and a task classifier labels the shape of the work (chat, extraction, code_edit, code_agent, tool_agent, reasoning_math, long_context, vision, …).
2. Models that cannot satisfy the request — no tool calling, no vision, too small a context window or max-output, incompatible structured output — are removed before anything is scored.
3. Survivors are ranked on task quality, capability, estimated cost, speed and reliability; the full ranked list is kept as the fallback chain for timeouts and 5xx.
Curated primaries on the default `auto` profile:
- **SIMPLE** — factual lookups, greetings, translations → gemini-2.5-flash
- **MEDIUM** — summaries, explanations, data extraction → kimi-k2.7
- **COMPLEX** — code generation, multi-step analysis → gemini-3.1-pro
- **REASONING** — proofs, formal logic, multi-step math → grok-4-1-fast-reasoning
Turns that actually need their attached tools switch to agent-tuned tiers (gpt-4o-mini → kimi-k2.7 → claude-sonnet-4.6). `eco` opens on the free tier (nemotron-3.5-lightning); `premium` climbs to gpt-5.3-codex / claude-fable-5 / claude-sonnet-4.6. Full chains: [docs/routing-profiles.md](../../docs/routing-profiles.md).
Per-tier
savings percentages are deliberately not quoted here: the published figure is
blended across a stated workload mix, and a per-tier number invites comparing
it against a baseline nobody wrote down. See
[savings-mix.json](https://github.com/BlockRunAI/blockrun/blob/main/src/brand/savings-mix.json).
## Available Models
<!-- br:models.chatVisible -->90<!-- /br:models.chatVisible --> models including: gpt-6-astra [GPT-6 flagship, 1M ctx], gpt-6-sol, gpt-6.1-sol [refreshed Sol], gpt-6-luna [fast], gpt-5.6-terra [balanced, stable default], gpt-5.6-sol [flagship], gpt-5.6-luna [cost-efficient], gpt-5.6-sol/terra/luna-pro [pro reasoning tiers], gpt-5.5, gpt-5.5-pro [max compute], chat-latest [ChatGPT Instant], gpt-5.4, gpt-5.1, gpt-4o, o3, claude-fable-5.1, claude-fable-5, claude-opus-5.5 [newest Opus], claude-opus-5 [1M ctx], claude-opus-4.8, claude-opus-4.7, claude-opus-4.5, claude-sonnet-5.5, claude-sonnet-5, claude-sonnet-4.6, gemini-3.1-pro, gemini-3.6-flash, gemini-3.5-flash-lite, gemini-2.5-flash, deepseek-v4-pro, deepseek-chat, grok-4.7, grok-4.6, grok-4.5, grok-4.3, grok-build-0.1, glm-5.3 [Z.AI flagship, 1M ctx], glm-5.3-flash [natively multimodal, 1M ctx], glm-5.2, minimax-m3, kimi-k3 [1M ctx flagship], kimi-k2.7, qwen3.8-max [Qwen 3.8 flagship, 1M ctx], qwen3.7-max, qwen3.7-plus, qwen3.7-flash, mimo [Xiaomi MiMo-V2.5 Pro, 1M ctx], mistral-large-4 [Mistral flagship, 512K ctx, vision], hy4-preview [Tencent, 770B MoE], laguna-s-2.1 [Poolside, coding], and <!-- br:models.free -->7<!-- /br:models.free --> free open-weight models (nemotron-3.5-lightning [1M ctx, thinking mode — the free default], nemotron-3-ultra-550b [550B/55B MoE, 1M ctx — the largest free model], nemotron-3-nano-omni-30b-a3b-reasoning [256K ctx], llama-3.2-11b-vision [Meta Llama], north-mini-code [Cohere, coding, sub-second], laguna-xs-2.1 [Poolside, coding, ~161 tok/s], muse-glimmer-30b [Meta Muse, reasoning]). The free tier is text-only — image turns route to a paid vision model, because the two free models the catalogs advertise as vision-capable fail a solid-colour probe on both chains and do so with an HTTP 200.
## Model Prices — Look Them Up Live
Model prices move (vendors reprice, promos end), so this skill carries no model price table. Read the live catalog before you budget a call:
```bash
curl -s https://blockrun.ai/api/v1/models # Base
curl -s https://sol.blockrun.ai/api/v1/models # Solana
curl -s https://arc.blockrun.ai/api/v1/models # Arc (same catalog as Base)
```
Each entry in `data[]` has an `id`, a `billing_mode` and a `pricing` object. All amounts are USD.
| `billing_mode` | `pricing` fields | Estimated cost of one call |
| ---------------- | ------------------------------------------------------------------------------------------------------ | -------------------------------------------------------------------------------------------------------------- |
| `paid` (chat) | `input`, `output` per 1M tokens; optional `cache_read` / `cache_write` per 1M; optional `long_context` | `(input_tokens × input + output_tokens × output) / 1,000,000`, minimum $0.001, plus the $0.001 transaction fee |
| `free` | `input: 0`, `output: 0` | $0 — no payment |
| `per_image` | `per_image`, `transaction_fee` | `per_image × images + transaction_fee` |
| `per_second` | `per_second`, `per_call_surcharge`, `transaction_fee`, `default_duration_seconds` | `per_second × seconds + per_call_surcharge + transaction_fee` |
| `per_track` | `per_track`, `transaction_fee` | `per_track + transaction_fee` |
| `per_character` | `per_1k_chars`, `transaction_fee` | `per_1k_chars × characters / 1,000 + transaction_fee` |
| `per_generation` | `per_generation`, `transaction_fee` | `per_generation + transaction_fee` |
- **Cached prompt tokens** bill at `cache_read` instead of `input` (and cache writes at `cache_write`) when the model lists them.
- **`long_context`** is a list of tiers. When the prompt is larger than a tier's `threshold` (or equal to it, if `inclusive` is true), that tier's `input` / `output` / `cache_*` rates apply to the **whole** request.
- Image size/quality and video resolution can raise the media price above the listed base rate.
- The figure above is an estimate. The exact amount is quoted in the HTTP 402 challenge before anything is signed, so a call never settles above its quote.
- Human-readable rate card: https://blockrun.ai/pricing
## Built-in Agent Tools
In addition to LLM routing, ClawRouter exposes BlockRun's x402-gated data APIs as ready-to-use OpenClaw tools. Every tool is paid from the same USDC wallet — no extra setup, no extra API keys.
### Market Data
Realtime prices and historical OHLC across every asset class. The agent should call these directly instead of scraping finance sites.
| Tool | Coverage | Price |
| -------------------------- | ------------------------------------------------------------------------------- | ------------- |
| `blockrun_stock_price` | 12 global markets: US (NYSE/Nasdaq), HK, JP, KR, UK, DE, FR, NL, IE, LU, CN, CA | $0.001 / call |
| `blockrun_stock_history` | OHLC bars at 1/5/15/60/240-min or D/W/M resolution | $0.001 / call |
| `blockrun_stock_list` | Ticker lookup / company-name search per market | Free |
| `blockrun_crypto_price` | BTC-USD, ETH-USD, SOL-USD, and more | Free |
| `blockrun_fx_price` | EUR-USD, GBP-USD, JPY-USD, and more | Free |
| `blockrun_commodity_price` | XAU-USD (gold), XAG-USD (silver), XPT-USD (platinum) | Free |
### Image & Video Generation
| Tool | Purpose | Price |
| --------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | ---------------------------------------------------------------- |
| `blockrun_image_generation` | 12 image models — GPT Image 1/2/2.5 (Flare, Sunburst), Nano Banana / 2 / Pro, Seedream 5 Pro, Grok Imagine / 2.0 / Pro, CogView-4 | per model — see [Model Prices](#model-prices--look-them-up-live) |
| `blockrun_image_edit` | Edit / inpaint existing image (GPT Image 1/2, 2.5 Sunburst, Nano Banana; `model` param) | per model — see [Model Prices](#model-prices--look-them-up-live) |
| `blockrun_video_generation` | Grok Imagine + ByteDance Seedance (1.5-pro / 2.0-fast / 2.0) at 720p with synced audio (t2v default), 5–10s, + OpenAI Sora 2 via Azure (`azure/sora-2`, 4/8/12s, t2v + i2v, rejects human faces in reference images). Seedance is token-priced upstream (~20,256 tokens/sec at the 720p+audio default — 2× the prior 480p rate); `image_url` (image-to-video) costs the same as t2v since 2026-06-01. Seedance 2.0 variants accept optional `real_face_asset_id` (`ta_…`) for BytePlus RealFace character-consistency — mutually exclusive with `image_url`. | per model — see [Model Prices](#model-prices--look-them-up-live) |
### Phone & Voice (Twilio + Bland.ai)
Verify phone numbers and place AI-powered outbound voice calls. **Real-world side effects** — only call `blockrun_voice_call` when the user has explicitly asked to place a call. Server enforces an emergency-number blocklist; ClawRouter does not.
| Tool | Purpose | Price |
| -------------------------------- | -------------------------------------------------------------------- | ---------- |
| `blockrun_phone_lookup` | Carrier + line type (mobile/landline/voip) for any E.164 number | $0.01 |
| `blockrun_phone_lookup_fraud` | SIM-swap + call-forwarding fraud signals (use before SMS-code flows) | $0.05 |
| `blockrun_phone_numbers_buy` | Provision a US/CA number bound to this wallet (30-day lease) | $5.00 |
| `blockrun_phone_numbers_renew` | Extend a wallet-owned number's lease 30 days | $5.00 |
| `blockrun_phone_numbers_list` | List numbers the wallet currently owns + expiry timestamps | $0.001 |
| `blockrun_phone_numbers_release` | Release a wallet-owned number back to the pool | free |
| `blockrun_voice_call` | AI outbound call via Bland.ai — up to 30 min, transcript + recording | $0.54 flat |
| `blockrun_voice_status` | Poll a call's status / transcript / recording | free |
Voice calls are **fire-and-forget**: the POST returns a `call_id` + `poll_url` immediately, the call itself runs in the cloud for up to 30 minutes. Poll `blockrun_voice_status` every 10–30s while in_progress to retrieve the transcript. Slash command: `/cr-call +1<E.164> "<task>" [--voice nat] [--max-duration 5]`. CLI: `clawrouter phone numbers list/buy/renew/release` and `clawrouter phone lookup/fraud <+E.164>`. See the `phone` skill for the full call-flow reference.
### Prediction Markets (Predexon)
Full prediction-market toolbox spanning **Polymarket, Kalshi, Limitless, Opinion, Predict.Fun, dFlow** + Binance for crypto candles. **57 endpoints (Predexon v2) exposed as 9 agent tools** (8 named ergonomic wrappers + 1 catch-all):
- **Markets & trading** — events, markets list per venue, cross-venue search (`markets/search`), orderbooks, candlesticks (per-market and per-token), trades, positions, volume charts.
- **Leaderboard & smart money** — global + per-market leaderboards, smart-money positioning, top holders, smart-activity feed.
- **Wallet analytics** — full wallet profile, P&L time series, per-market breakdown, similar-wallet discovery, batch profiles, AND/OR filters.
- **UMA oracle + wallet identity** — UMA optimistic-oracle resolution status (`uma/markets`, `uma/market/{conditionId}`); wallet identity labels (ENS / Lens / exchange / risk tags), bulk identity, on-chain cluster discovery.
| Tool | Coverage | Price |
| ------------------------------------ | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | ---------------------- |
| `blockrun_predexon_events` | Live Polymarket events with current odds | $0.001 / call |
| `blockrun_predexon_markets` | Search Polymarket markets by keyword | $0.001 / call |
| `blockrun_predexon_leaderboard` | Top Polymarket traders ranked by profit | $0.001 / call |
| `blockrun_predexon_smart_money` | Smart-money positions on a specific market | $0.005 / call |
| `blockrun_predexon_smart_activity` | Markets where smart money is currently active | $0.005 / call |
| `blockrun_predexon_wallet` | Polymarket wallet profile (PnL, winrate, positions) | $0.005 / call |
| `blockrun_predexon_wallet_pnl` | Wallet P&L time series | $0.005 / call |
| `blockrun_predexon_matching_markets` | Polymarket ↔ Kalshi market pairs (arb compare) | $0.005 / call |
| `blockrun_predexon_endpoint_call` | Catch-all for the remaining 49 endpoints — orderbooks, candlesticks, top-holders, UMA oracle, wallet identity/cluster, Kalshi/Limitless/Opinion/Predict.Fun, dFlow, Binance Futures, cross-venue search, sports, canonical markets. Takes `path` + optional `method`/`query`/`body`. | $0.001 / $0.005 / call |
Pricing: `$0.001` per market-data call, `$0.005` per analytics / search / wallet call. See the `predexon` skill for the full endpoint reference.
### Prediction-Market Trading (Polymarket) — REAL MONEY
Beyond reading odds, ClawRouter can **place, manage, and redeem real bets** on
Polymarket (CLOB V2, Polygon) via the `blockrun_polymarket` tool. Unlike every
other tool here it is **not** an HTTP-proxy wrapper — it runs a local trading
engine that signs CLOB orders (EIP-712) with the SAME ClawRouter wallet that
pays for LLM calls, and posts them to Polymarket (through BlockRun's Tokyo
egress by default, so it works out of the box in geoblocked regions).
| Tool | What it does | Cost |
| --------------------- | ------------------------------------------------------------------------------------------------------------ | -------------------------- |
| `blockrun_polymarket` | `action:` setup / fund / buy / sell / cancel / orders / positions / redeem / withdraw. One multiplexed tool. | Free tool; bets spend pUSD |
- **Funds**: bets spend **pUSD** in a gasless Polygon deposit-wallet vault
derived from your key; `action:"fund"` moves your **Base USDC → pUSD** gaslessly
(x402, $0.01 fee, non-custodial). Winnings `withdraw` back to native USDC on Base.
- **Safety**: `confirm:true` is **hard-required** to place/sign anything (omit →
dry-run preview); per-order cap `POLYMARKET_MAX_BET_USD` (default $25) + optional
`POLYMARKET_MAX_SESSION_USD`. Bets do NOT draw from the x402 API budget.
- **Zero config**: no Polymarket account, API keys, or gas token — first `setup`
bootstraps everything from your wallet. Discover token IDs with the
`blockrun_predexon_*` data tools first. See the **`polymarket-trading` skill**
for the golden rules and end-to-end flow.
## Example Output
```
[ClawRouter] google/gemini-2.5-flash (SIMPLE, rules, confidence=0.92)
Cost: $0.0025 | Baseline: $0.308 | Saved: 99.2%
```