What is predictioncouncil?

A research tool that asks four AI models to estimate the probability of events traded on Polymarket, a market where people buy and sell predictions about real-world outcomes. It compares their views, reviews them, and produces a combined verdict.

Use it to find a Polymarket event, compare its market odds with AI estimates, and get a council report about the likelihood of an outcome.

It gives you several model opinions in one report instead of making you inspect them separately. Public results are later compared with what actually happened, so you can review how well the predictions were calibrated.

predictioncouncil — Skill for Claude Code and Codex by daveb1392

Install

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.

Any agent

npx skills add daveb1392/predictioncouncil-mcp --skill predictioncouncil

Clone the repo

git clone --depth 1 https://github.com/daveb1392/predictioncouncil-mcp

Made for: Claude Code, Codex.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/daveb1392/predictioncouncil-mcp/predictioncouncil/github.svg)](https://agentmods.dev/skills/daveb1392/predictioncouncil-mcp/predictioncouncil)

Your own site

<a href="https://agentmods.dev/skills/daveb1392/predictioncouncil-mcp/predictioncouncil"><img src="https://agentmods.dev/badge/skills/daveb1392/predictioncouncil-mcp/predictioncouncil/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.

Your own site · 80×15

<a href="https://agentmods.dev/skills/daveb1392/predictioncouncil-mcp/predictioncouncil"><img src="https://agentmods.dev/badge/skills/daveb1392/predictioncouncil-mcp/predictioncouncil.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 88 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 639 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00088	$0.00639
Opus 5	$0.00044	$0.00319
Sonnet 5	$0.00018	$0.00128
Haiku 4.5	$0.00009	$0.00064

Measured 10d ago against content hash eeb0ea38511d, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

predictioncouncil 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 10d 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.

skills/predictioncouncil/SKILL.md · 51 lines

How it starts

The opening of the file, as written. The whole thing — 51 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Prediction Council

Prediction Council (predictioncouncil.com) runs a 4-model AI council on any Polymarket event: independent opinions → cross-review → chairman synthesis. Every verdict is later graded against the real outcome on a public track record, so its calibration is auditable (GET /api/stats/track-record).

Preferred access: MCP tools

If polyinsight_* MCP tools are available, use them. Otherwise use the REST API at https://backend-production-092f.up.railway.app with header X-API-Key: $POLYINSIGHT_API_KEY.

Workflow (always in this order)

Find the event — polyinsight_search_events (query, category). Free.
Check before spending — polyinsight_check_report(event_id). Returns the current dynamic quote (cost_credits) and whether a same-day cached report exists (half price, or free if already purchased). NEVER skip this: pricing is dynamic and running a report spends real credits.
Confirm with the user if the run costs credits and they haven't already approved spending in this conversation.
Run — polyinsight_run_council_report(event_id). Takes 30–120s; the tool polls to completion. Rate limit 5 runs/hour; failed runs auto-refund.
Read the verdict — key fields:
- most_likely_outcome + most_likely_probability (the headline answer)
- market_probabilities (the council's full distribution)
- top_bets (ranked suggestions with payout math; kind:"edge" = mispricing vs kind:"pick" = fair-price favorite)
- provider_failures / degraded (which models actually participated)

Interpreting results honestly

The council can be wrong — cite its public hit rate from /api/stats/track-record when presenting confidence.
SKIP on the queried market is common on efficiently priced events; the distribution is still informative.
This is analysis, not financial advice; say so when relaying bet suggestions.

Cost economics

Fresh report: dynamic quote (typically a few hundred credits ≈ $4–6).
Same-day cached report: half. Already purchased: free.
1 free report/week per account; buy packs or a metered API subscription ($3 fresh / $1.50 cached) at predictioncouncil.com.

Read the full file on GitHub · 51 lines

Changes

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.

10d ago First seen · 51 lines · 88 tokens per session scan A eeb0ea38511d

Subscribe to this mod's changes

predictioncouncil is a skill published in the GitHub repository daveb1392/predictioncouncil-mcp (0 stars, last pushed 1mo ago), licensed MIT. It adds 88 tokens to every session and 639 once invoked, about $0.0004 per session on Opus 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 2026-08-31.

Finance and fintech Research

sector-rotation

An analysis framework for comparing industries in the Chinese A-share stock market, using business conditions, price momentum, valuation, and money flows. It produces rankings and higher- or lower-allocation suggestions.

HKUDS/Vibe-Trading · 39 tokens

strategy-pivot-designer

Detect backtest iteration stagnation and generate structurally different strategy pivot proposals when parameter tuning reaches a local optimum.

tradermonty/claude-trading-skills · 28 tokens

twitter-reader

Read Twitter/X for financial research using opencli (read-only). Use this skill whenever the user wants to read their Twitter feed, search for financial tweets, view bookmarks, look up user profiles, or gather market sentiment from Twitter/X. Triggers include: "check my feed", "search Twitter for", "show my…

himself65/finance-skills · 161 tokens

predictioncouncil