Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add hive-intel/hive-sdk --skill hivegit clone --depth 1 https://github.com/hive-intel/hive-sdkWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/hive-intel/hive-sdk/hive)<a href="https://agentmods.dev/skills/hive-intel/hive-sdk/hive"><img src="https://agentmods.dev/badge/skills/hive-intel/hive-sdk/hive/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/hive-intel/hive-sdk/hive"><img src="https://agentmods.dev/badge/skills/hive-intel/hive-sdk/hive.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00165 | $0.05196 |
| Opus 5.5 | $0.00066 | $0.02078 |
| Sonnet 5.5 | $0.00033 | $0.01039 |
| Haiku 4.5 | $0.00016 | $0.00520 |
Grade A, and why
hive scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 25d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 407 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Hive: one skill, all crypto intelligence
Hive answers live crypto questions from more than a dozen data providers through one MCP server and one CLI, and every answer carries a receipt: provider, fetch time, cache state, runtime status, and digests of the input and result. Three things Hive has that most crypto data tools do not:
- History that venues no longer serve. Perp funding settlements for one coin across up to 33 CEX and DEX venues since 2023-11, plus basis, open interest and liquidations since 2025 and long/short ratios since 2026-03.
- Cross-venue views. One call compares funding or basis across venues instead of one exchange at a time.
- Provenance you can cite. The
_hiveblock on every material response is the receipt. Copy it; never invent one.
Which transport
Detect the host once per session:
- Hive tools are present in your tool list (
search_tools,invoke_api_endpoint,get_token_price, ...): use the MCP path. Do not shell out to the CLI for data the MCP path already serves. - No Hive tools in the tool list, but you have a shell: use the CLI path
(
hive ...). - The user says "use the CLI" or "use MCP", or the environment sets
HIVE_TRANSPORT=cli|mcp: that wins.
Both paths return the same receipts and the same error codes.
Setup
One incantation on every surface. It registers the hosted MCP endpoint in every detected client, copies the skill packs, and skips the post-install probe so it also works in non-interactive shells:
npx -y -p hive-intelligence@latest hive init --all --skip-verify
Never run bare hive init from an agent: it is interactive and exits 1 in a
non-TTY shell. Add --browser only when the user asked to sign in.
Hosted MCP endpoint: https://mcp.hiveintelligence.xyz/mcp (Streamable HTTP,
OAuth in the client browser; headless backends send
Authorization: Bearer <HIVE_API_KEY> from secret storage).
Version check, once per session on the CLI path:
npx -y -p hive-intelligence@latest hive --version
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 25d ago Changed · +2 lines 66c067274671
- 1mo ago Changed · +22 lines cb7c4e05195e
- 1mo ago First seen · 383 lines · 165 tokens per session scan A cc19fc875d9b
hive is a skill published in the GitHub repository hive-intel/hive-sdk (20 stars, last pushed 3d ago), licensed MIT. It adds 165 tokens to every session and 5,196 once invoked, about $0.0007 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-17.
Other skills, from other repositories
pump-swarm
Coordinated multi-wallet trading on Pump.fun.
trading-solana
Trade tokens on Solana DEXes - Jupiter, Raydium, Orca, Meteora, Pump.fun.
copy-trading
Automatically copy trades from successful wallets on Polymarket and crypto.
trading-evm
Trade tokens on EVM chains - Uniswap V3, 1inch on Ethereum, Arbitrum, Optimism, Base, Polygon.
whale-tracking
Monitor whale trades on Polymarket and crypto chains (Solana, ETH, Polygon, ARB, Base, OP).
acp
Enable agent-to-agent commerce with on-chain escrow, cryptographic agreements, and service discovery.