Axion Forecasting for Openclaw

An advanced AI-driven research and forecasting skill that runs multi-agent pipelines to generate graded, evidence-backed probabilistic estimates.

eternis
v1.0.2
Jun 26, 2026
1
376
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install axion

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install axion using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is Axion Forecasting?

Axion Forecasting is a high-performance skill designed to calculate the precise probabilities of future events, macroeconomic shifts, business deals, and technological trends. By integrating this skill with Openclaw Skills, developers can deploy a coordinated swarm of specialized research agents. These agents scrape live web search indices, evaluate market metrics, dissect SEC filings, and formulate Fermi estimates to output deep-dive quantitative forecasts.

Rather than generating simple heuristics, Axion operates on an asynchronous model. It crawls multiple sources, parses real-world evidence, and generates highly detailed probabilistic metrics. This structured output is ideal for risk analysis, strategic planning, financial models, and any decision-making process that requires quantifiable risk assessment rather than static text predictions.

Axion Forecasting Use Cases

  • Evaluating geopolitical outcomes, macroeconomic changes, or interest rate movements (e.g., forecasting Federal Reserve rate cuts).
  • Assessing business transaction success rates, corporate acquisitions, and competitive product launches.
  • Automating financial market analysis and investment research through real-time SEC filing evaluation.
  • Formulating structural Fermi estimates and probability profiles for complex software or hardware releases.

How Axion Forecasting Works

  1. Initialize Request: The skill launches a new forecasting run by issuing an asynchronous POST request to the Axion API with a prompt and an allocated effort level.
  2. Multi-Agent Research: An autonomous orchestrator deploys agents to extract and analyze data from the web, SEC databases, and market feeds.
  3. Asynchronous Polling: The skill performs periodic non-blocking HTTP GET requests to monitor the processing state.
  4. Deliver Probabilistic Data: Once execution completes, the skill compiles and presents calculated probabilities, upper/lower confidence boundaries, and structured reasoning trails.

Axion Forecasting Setup

To deploy the Axion forecasting skill, complete the following setup steps:

  1. Sign up for an account at Axion Eternis and generate an API key on the API Keys dashboard.
  2. Ensure your pre-paid credit balance is active, as forecasting tasks are billed per-run separate from plans.
  3. Export your credentials to the host or sandboxed environment using the AXION_API_KEY variable:
export AXION_API_KEY="your_axion_api_token_here"

The forecasting agents will automatically detect this environment variable to sign outgoing HTTPS requests.

Axion Forecasting Data Schema & Taxonomy

The Axion forecasting skill uses structured JSON payloads to manage query limits and detail results. Below is the primary schema layout:

Input Configurations

Parameter Data Type Required Description
input String Yes The target query or future event to analyze (e.g., 'Will the Fed cut rates in June 2026?').
effort String No Computation depth setting: low, medium, or high.
max_forecasts Integer No Range 1-10. Spawns multiple parallel forecast evaluations to broaden predictions.

Response Properties

When a forecast transitions to a completed state, the skill returns an array of structured objects featuring:

  • probability: Float value between 0 and 1 expressing the calculated event probability.
  • confidence_lower / confidence_upper: Float values setting the confidence intervals of the prediction.
  • forecast_text: Descriptive technical evaluation summarizing the predicted outcome.
  • reasoning: The core analytical logical steps and cited sources that validate the calculation.
  • resolution_date: Expected calendar date when the target scenario will resolve.

Axion Forecasting Advanced Features

  • Multi-Agent Orchestration Swarms: Automatically utilizes diverse agent protocols specialized in web crawling, SEC filing reading, and mathematical estimation.
  • Configurable Resource Allocation: Scales computation settings (low/medium/high) and parallel streams (up to 10 concurrent runs) to match search depth criteria.
  • Error Resilience and Network Robustness: Implements status-code pre-parsing safeguards to elegantly manage transient 5xx ingress drops using progressive backoff retries.
  • Sandbox Environment Compatibility: Securely mounts credentials inside container environments without exposing operational secrets within logging consoles.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Requires
Bins curl
Github Stars: 0
forks: 0

Featured*