agentcanary_skill

This skill provides macro regime context, risk assessment, and market structure insights to guide position sizing and strategic decisions.

Python

2.5k

GitHub Stars

3

Bundled Files

2 months ago

Catalog Refreshed

3 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill openclaw/skills --skill agentcanary

_meta.json274 B
endpoints.md24.4 KB
SKILL.md6.5 KB

Overview

This skill is a market intelligence API designed for AI agents that need macro regime context, risk scoring, and trading signals across crypto, stocks, FX, and commodities. It exposes 33 read-only endpoints delivering regime labels, composite risk gauges, technical signals (IGNITION/ACCUMULATION/DISTRIBUTION/CAPITULATION), whale alerts, orderbook analytics, news sentiment, and treasury tracking. The API covers 1,181 assets and aggregates 250+ sources to return actionable intelligence rather than raw data.

How this skill works

Agents make HTTP GET calls to specific endpoints to fetch JSON intelligence: macro snapshots, multi-timeframe signal states, technical indicator panels, orderbook depth, and news with FinBERT sentiment. The skill classifies regimes and states (not price predictions), scores risk, surfaces whale and funding-arbitrage opportunities, and returns AI-generated daily market reports. It is API-only, read-only, and requires no local execution or secrets embedded in prompts.

When to use it

Provide macro regime context to suppress or allow trading strategies (risk-on vs risk-off).
Confirm multi-timeframe signal state before entering or sizing a position.
Monitor whale transactions and funding-rate arbitrage opportunities in real time.
Detect orderbook liquidity shifts and persistent walls before large trade execution.
Ingest breaking news and FinBERT sentiment to interrupt or adjust automated decisions.

Best practices

Call /macro-snapshot/regime every 4–6 hours to keep regime context current.
Use /signal-state and /cointa for confirmation, not as sole trade drivers.
Poll whale alerts and breaking news every 15–30 minutes for timely interrupts.
Treat severity flags as interrupts—combine with your execution and risk logic.
Avoid embedding API keys or secrets in prompts; use secure wallet-based auth when available.

Example use cases

An agent suppresses opening new positions when /macro-snapshot/regime returns Risk-Off.
Live monitoring pipeline triggers reduced position sizes when whale-alerts show large outflows.
Execution agent checks /orderbook/depth and wall-persistence before slicing a large order.
Portfolio manager uses /coin-rsi/statistics to rebalance baskets of oversold assets.
News-driven agent pauses or hedges positions when /news/breaking shows high-negative FinBERT sentiment.

FAQ

No. The API is read-only and provides intelligence; execution and order placement remain the agent's responsibility.

Can the API predict prices?

No. It classifies regimes, signals, and risk but does not produce price forecasts or guaranteed returns.

How often should I poll for updates?

Follow the signal cadence: macro every 4–6h, signal states hourly to daily, whale/news every 15–30 min, and prices as needed.