Syntic
← All skills
Strategy R&DFreeCommunityv1.0.0

trade-hypothesis-ideator

Use when trade logs, journal entries, or market observations suggest a potential edge and the user wants that turned into falsifiable, ranked trading hypotheses with kill criteria.

About this skill

Use when trade logs, journal entries, or market observations suggest a potential edge and the user wants that turned into falsifiable, ranked trading hypotheses with kill criteria.

SKILL.md

trade-hypothesis-ideator.SKILL.md

Place it in your Syntic skills folder. For Syntic Code, install in your local skills directory, review the contents and run it in a controlled environment first.

SKILL.md
---
name: trade-hypothesis-ideator
description: Use when trade logs, journal entries, or market observations suggest a potential edge and the user wants that turned into falsifiable, ranked trading hypotheses with kill criteria.
category: Strategy R&D
version: 1.0.0
tools: []
---

# Trade Hypothesis Ideator

Generate 1-5 structured hypothesis cards from trade logs, journal entries, and market observations, critique and rank them, then optionally turn the winners into exportable strategy specs.

## When to Use

- After gathering trade logs, journal entries, or market observations that suggest a potential edge
- There's a bundle of evidence (trade history, journal snippets, market data, observations) and the goal is falsifiable hypotheses, not vague ideas
- To bridge qualitative observations into quantitative experiment designs
- Before committing capital, to validate a new strategy idea against explicit kill criteria

## Methodology

1. **Normalize and extract evidence** from whatever mix of trade logs, journal snippets, market data, and observations is available into a structured summary of recurring patterns and quantifiable signal.
2. **Generate hypotheses** — draft 1-5 cards, each a specific, falsifiable claim about market behavior (not "momentum works" but a precise, testable mechanism with conditions).
3. **Critique** each: is the evidence sample large enough? Is there a plausible causal mechanism, or just pattern-matched coincidence? What would disprove it?
4. **Rank and assign a verdict**: `pursue` (evidence and mechanism strong enough to design an experiment), `revise` (promising but needs sharpening), or `discard` (too thin or implausible).
5. **Export (optional).** For `pursue` cards, produce an experiment design: entry/exit logic sketch, sample-size requirement, and explicit kill criteria — conditions under which the hypothesis should be abandoned before a full backtest. This becomes a strategy spec ready for a strategy-design/backtesting pipeline, with provenance tracing it back to the source evidence.

## Output

Per hypothesis: the falsifiable claim, supporting evidence excerpts, verdict, rationale, and — for `pursue` cards — the experiment design with kill criteria. Present as a ranked list in chat; include the strategy spec and provenance for exported cards.

Install

Syntic App

  1. 1Create a dedicated folder for this skill in your local skills library.
  2. 2Place SKILL.md into that folder.
  3. 3Restart Syntic and invoke this skill on matching tasks.

Syntic Code (CLI)

  1. 1Save SKILL.md in your local Syntic Code skills directory.
  2. 2Keep related files in the same skill folder.
  3. 3Run in a safe environment and validate outputs.