gentech-blockrun
Skill389 starsChanged 23 days agoMITView on GitHub
GenTech Labs' integration patterns for BlockRun MCP from Hermes Agent. Covers daily usage patterns, cost-optimized workflows, multi-tool pipelines, and reliable error handling for BlockRun's full toolset.
In this repository18 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- blockrun-debugSkill
- blockrun-setupSkill
- blockrunSkill
- blockrun-upgradeSkill
- crypto-dataSkill
- decideSkill
- exa-researchSkill
- image-promptingSkill
- modalSkill
- phoneSkill
- polymarket-tradingSkill
- prediction-marketsSkill
- rpcSkill
- searchSkill
- signal-to-trade-demoSkill
- surfSkill
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 2 are distinctive, and 4 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 88
Start with the free tools, then pay only for what they cannot answer.
- Distinctive: line 109
Use Exa (neural search) + BlockRun Chat (second opinion) for thorough research.
- Very specific: line 123
Cost-conscious tip: Use mode: "free" on blockrun_chat for zero-cost generation (NVIDIA models).
- Very specific: line 127
Use blockrun_wallet + agent_id for multi-agent cost tracking.
- Very specific: line 277
Token prices: No need to cache aggressively — crypto/FX/commodity quotes via blockrun_price are FREE.
- Very specific: line 299
Note: Every paid media tool (image, video, music, speech, realface) pays on either chain.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
GenTech Labs — BlockRun Integration Patterns
Real-world patterns from running BlockRun MCP in production across 32 cron jobs and daily agent operations.
Getting Started
Initial Wallet Setup
# Set up BlockRun wallet (one-time)
blockrun_wallet(action="setup")
# → Follow the QR to fund the wallet with USDCBudget Controls
# Set a global budget cap so spending never surprises you
blockrun_wallet(action="budget", budget_action="set", budget_amount=10.0)
# Spawn child agents with delegate budgets
blockrun_wallet(action="delegate", agent_id="researcher", agent_limit=2.0)
blockrun_wallet(action="delegate", agent_id="data-poller", agent_limit=0.50)Check Balance
blockrun_wallet(action="status")
# → Two wallets: Base (also pays on Arc) + Solana. Active chain shown. Check before expensive runs.Daily Usage Patterns
Pattern 1: Regular Price Checks (FREE)
Crypto, FX and commodity quotes cost nothing — blockrun_price is free for those
categories (stocks/usstock quotes are not served since 2026-09-05 — the gateway 501s before payment; only the ticker catalog works). Use them liberally.
(blockrun_surf, which used to charge $0.0085 for the same quote, was removed in 0.49.0 — the gateway 410s every Surf path since 2026-09-06.)
# Single price
blockrun_price(action="price", category="crypto", symbol="BTC-USD")
# Batch of prices
blockrun_price(action="price", category="crypto", symbol="ETH-USD")
blockrun_price(action="price", category="crypto", symbol="SOL-USD")
# Free — discover available symbols
blockrun_price(action="list", category="crypto", query="sol")Cost: $0 — crypto/FX/commodity price and list calls are both free. category:"stocks" price/history currently return 501 (equity quotes withdrawn 2026-09-05, nothing charged); its list catalog is free.
Pattern 2: Token Research Pipeline (~$0.018)
Start with the free tools, then pay only for what they cannot answer.
# 1. Free — DEX liquidity + volume (free endpoint)
blockrun_dex({ query: "SOL" })
# 2. FREE — CEX price (crypto category is free)
blockrun_price(action="price", category="crypto", symbol="SOL-USD")
# 3. $0.0060 each — Protocol TVL
blockrun_defi({ path: "protocols" })
blockrun_defi({ path: "protocol/jupiter" })
# 4. $0.0060 — Chain TVL
blockrun_defi({ path: "chains" })Total: ~$0.018 per full token analysis (2 free calls + 3 x $0.0060 DefiLlama). Replace CoinGecko/CMC tabs entirely.
Pattern 3: Research + Synthesis (~$0.011 + chat)
Use Exa (neural search) + BlockRun Chat (second opinion) for thorough research.
# 1. $0.0110 — Neural web search with Exa
blockrun_exa({ path: "search", body: {
query: "latest AI agent infrastructure developments 2026",
numResults: 10,
category: "research paper"
}})
# 2. per-token — Get second opinion from GLM-5 (excellent for technical details)
blockrun_chat({ mode: "glm", message: "Summarize the key trends..." })Cost-conscious tip: Use mode: "free" on blockrun_chat for zero-cost generation (NVIDIA models).
Pattern 4: Agent-to-Agent Billing
Use blockrun_wallet + agent_id for multi-agent cost tracking.
# Allocate budget per agent group
blockrun_wallet(action="delegate", agent_id="research", agent_limit=2.0)
blockrun_wallet(action="delegate", agent_id="content", agent_limit=1.0)
# Pass agent_id on every call
blockrun_defi({ path: "protocols", agent_id: "research" })
blockrun_search({ body: { query: "latest news", sources: ["web"] }, agent_id: "research" })
# Audit at end of day
blockrun_wallet(action="report")Multi-Tool Pipelines
DeFi Intelligence Pipeline
# 1. TVL & protocol health ($0.0060)
blockrun_defi({ path: "protocol/aave-v3" })
# 2. Yield pools ($0.0060)
blockrun_defi({ path: "yields" })
# 3. Token price (FREE — crypto category)
blockrun_price(action="price", category="crypto", symbol="AAVE-USD")
# 4. DEX activity (free)
blockrun_dex({ token: "0x7Fc66500c84A76Ad7e9c93437bFc5Ac33E2DDaE9" })Portfolio Health Check
# 1. Current holdings (free — local data or blockrun_rpc)
blockrun_rpc({ network: "base", method: "eth_getBalance", params: ["0xabc...", "latest"] })
# 2. Token prices (FREE — crypto category)
blockrun_price(action="price", category="crypto", symbol="ETH-USD")
blockrun_price(action="price", category="crypto", symbol="USDC-USD")
# 3. Prediction market positions ($0.0085)
blockrun_markets({ path: "polymarket/positions" })Content Generation Pipeline
# 1. Research ($0.0110)
blockrun_exa({ path: "search", body: { query: "agentic commerce trends", numResults: 5 }})
# 2. Image generation for post ($0.01675 cheapest to $0.106 quality)
blockrun_image({ prompt: "AI agent futuristic dashboard with neon grids" })
# 3. Voiceover for video ($0.0535/1k chars)
blockrun_speech({ input: "Your narration text here...", voice: "sarah" })
# 4. Music track ($0.1585)
blockrun_music({ prompt: "upbeat synthwave background" })
# 5. Short video clip ($0.421)
blockrun_video({ prompt: "animated data visualization", duration_seconds: 8 })Cost Optimization Strategies
Tier Mapping
| Call Type | Price | Frequency |
blockrun_price (list) | FREE | daily discovery |
blockrun_dex | FREE | unlimited |
blockrun_rpc | $0.0030 | on-chain reads (batch: $0.002/element + $0.001) |
blockrun_wallet (status/report) | FREE | before every session |
blockrun_price (quote) | FREE | crypto/FX/commodity; stocks quotes not served (501 since 2026-09-05) |
blockrun_defi | $0.0060 | protocol/chain analysis ($0.0020 for prices/*) |
blockrun_chat (free mode) | $0 | NVIDIA-backed chat |
blockrun_chat (glm mode) | per-token | Zhipu GLM-5 coding — billed on tokens used, not a flat rate |
blockrun_exa (search) | $0.0110 | deep research (contents: $0.002/URL + $0.001) |
blockrun_speech | per model — live pricing at /api/v1/models | TTS |
blockrun_image | per model — live pricing at /api/v1/models | image generation |
blockrun_music | per model — live pricing at /api/v1/models | music tracks |
blockrun_video | per model — live pricing at /api/v1/models | short video clips (per-second × duration) |
blockrun_markets | $0.0085 | prediction market data, wallet analytics (flat) |
blockrun_modal | $0.0020–$192 | remote sandbox execution — timeout drives the price, see the modal skill |
blockrun_phone | $0.0110–$5.001 | lookups; number buy/renew is $5.001 |
Model-priced tools (chat beyond the free tier, speech, image, music, video) carry no number here because their prices move: read the model's pricing object from GET https://blockrun.ai/api/v1/models (Solana: https://sol.blockrun.ai/api/v1/models) — input/output per 1M tokens for chat, per_1k_chars for speech, per_image, per_track, or per_second for media, each plus transaction_fee. The HTTP 402 challenge quotes the exact amount before anything is signed.
Daily Budget Formula (GenTech's Setup)
# ~$0.30/day typical
Daily budget:
Free tools: unlimited (price quotes, dex, list, wallet status)
Crypto data: ~$0.10 (markets, flat $0.0085; defi $0.0060; surf is retired and free)
AI calls: ~$0.10 (chat free mode, occasional glm)
Media: ~$0.10 (occasional image/speech)
Total: ~$0.30/daySet a $5/week budget cap and never hit it with typical usage.
Error Handling Patterns
402 Payment Required
# When the wallet is FUNDED, x402 handles this automatically: the SDK
# signs the payment challenge and retries — no action needed.
#
# When the wallet is EMPTY, the 402 / insufficient-balance surfaces to you.
# Check status and fund FIRST — don't blindly retry the failing tool:
blockrun_wallet(action="status") # then action="setup" / "deposit"Insufficient Balance
# Before a batch of expensive calls:
balance = blockrun_wallet(action="status")
if float(balance["usdcBalance"]) < 0.50:
blockrun_wallet(action="deposit") # → opens Coinbase Onramp linkRate Limiting / Timeouts
# For async operations (video, music):
# They auto-poll until complete. On a timeout READ THE ERROR: on the Base
# wallet payment settles on completion, so a timeout is not charged and a
# retry is safe; Solana music and every account-rail media job are billed
# at submit, so the timed-out job IS charged, the error names it, and a
# retry bills a second job. Check blockrun_wallet action:"report" first.Caching Strategy
BlockRun calls are cheap but not free. Cache results that don't change minute-to-minute:
- Token prices: No need to cache aggressively — crypto/FX/commodity quotes via
blockrun_priceare FREE. - Protocol TVL: Cache for 1 hour (updates hourly)
- Wallet labels: Cache for 24 hours (rarely change)
- On-chain SQL results: Cache for 1 hour (most queries return the same answer)
GenTech's approach: Don't bother caching prices (crypto/FX/commodity quotes are FREE). Cache TVL and wallet labels aggressively.
Switching Chains
# Default is Base. Switch to Solana for Solana-native tools:
blockrun_wallet(action="chain", chain="solana")
# Switch back:
blockrun_wallet(action="chain", chain="base")
# Check which is active:
blockrun_wallet(action="chain")Note: Every paid media tool (image, video, music, speech, realface) pays on either chain. Base-only: blockrun_modal, blockrun_defi, claude-* chat with thinking (a plain claude-* call takes the compat path on Solana), and Polymarket funding. Solana works for price, wallet, dex, rpc, markets, etc.
Reference
- BlockRun MCP: https://github.com/BlockRunAI/blockrun-mcp
- BlockRun Docs: https://blockrun.ai/docs
- Models & Pricing: https://blockrun.ai/models
- Community: https://t.me/blockrunAI
Contributed by GenTech Labs, from real production usage of BlockRun MCP.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

