← Back to list
prediction-market-live-ops Checked
Use when building, backtesting, operating, or scaling automated trading on prediction markets (Ka…
8.1Overall rating
Installation
🤖 Install via AI
Copy the prompt below and send it to your AI assistant (e.g. Claude Code) — it will follow the instructions and install automatically.
Install the "prediction-market-live-ops" skill by following the instructions at https://skill123.me/install/prediction-market-live-ops.
⌨️ Command line install
Run in your terminal — downloads and installs to ~/.claude/skills/.
curl -fsSL https://skill123.me/install/prediction-market-live-ops.sh | bash
📦 Download ZIP
Download the zip and extract it into your skills directory (e.g. ~/.claude/skills/), then restart your session.
⬇ Download v1.0.0 · 3 KBAbout this skill
Overview
Use when building, backtesting, operating, or scaling automated trading on prediction markets (Kalshi or similar) — evidence-gated methodology, live-execution safety rails, and falsification protocols distilled from a real live-money campaign
Source
- Repo: https://github.com/agiprolabs/claude-trading-skills
- Path: skills/prediction-market-live-ops
Score breakdown
Trigger
7.5
Description states What (validating/executing/killing strategies), When (building/backtesting/operating/scaling automated PM trading), and venue scope; implicit When-Not via complement pointer to prediction-market-strategy; good keyword spread (live ops, safety rails, falsification).
Structure
7.3
Single-file skill, 106 lines, no offload needed; no redundant files; no progressive-disclosure structure because there is nothing to disclose progressively (neutral).
Workflow
7.6
Strong risk gating throughout (smoke-test 1-lot round trip, halt-file, drawdown breaker, equity floor, per-order caps) and excellent verification loops (paper/live EV ratio gates, reconcile-to-venue-fills, cohort attribution); but all guidance is prose doctrine — no scripts, checklists are embedded rather than procedural; stage progression is described but not operationalized.
Content
9.0
Very dense, imperative, incident-grounded writing with concrete costs ($84 runaway, 0-for-78) — high signal per line.
Engineering
9.0
Frontmatter valid name+description; no scripts or internal references; only external link is a GitHub repo.
Security
9.0
6.0 manifest: pure text doctrine — no network, file writes, credential access, or persistence. Capabilities match declared scope. Risk gating for live trading is the skill's core content (a compliance positive). Minor: references a third-party GitHub repo (agiprolabs/kalshi-stack) as reference implementation without vetting note; no explicit financial-risk disclaimer.
Operational doctrine for live-money prediction-market trading distilled from a real campaign: evidence-gated stage progression with pre-declared kill conditions, incident-ledger execution rules, drawdown breakers, and scaling ratchets. Single 106-line file, no scripts. Deductions: knowledge/doctrine orientation with no executable components or step-by-step procedure; boundary vs sibling skill only implicit; third-party GitHub repo promoted as reference implementation.