Shadow Market
PassAudited by ClawScan on May 12, 2026.
Overview
This appears to be a local prediction-scoring demo rather than a live market; it asks for no credentials or network access, but its financial-style claims and local log file should be treated cautiously.
Install only if you want a local experimental prediction-scoring concept. Do not treat its outputs as proven investment signals, and be aware that running the script can leave a local shadow_spine.jsonl-style log of events and predictions.
Findings (2)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
A user might mistake the generated shadow prices for actionable forecasting or investment intelligence.
The skill frames its outputs as alpha or prediction-market signals, while the included code only calculates synthetic scores from submitted probabilities. This is purpose-aligned but could be over-trusted.
Use when pricing undiscovered correlations, building AI-powered prediction markets, or extracting alpha from perception gaps between human and machine cognition.
Treat this as an experimental scoring/demo tool, not financial advice or a live prediction market, and require independent validation before acting on outputs.
Event names or prediction details entered into the tool may remain on disk after use.
Prediction captures, including event names and calculated prices, are appended to a local JSONL file. This is scoped and purpose-aligned, but it leaves persistent local data.
def __init__(self, spine_path: str = "shadow_spine.jsonl") ... with open(self.spine_path, "a") as f: f.write(json.dumps(entry) + "\n")
Avoid entering sensitive event names or private strategy details unless the output path and file retention are acceptable; delete the JSONL file when it is no longer needed.