long-short-pitch — Codex skill: what it does, how to install
Builds portfolio-manager-ready public equity pitches. It frames trade mechanics, catalyst timelines, and exit disciplines even under sparse context.
npx skills add openai/plugins --skill long-short-pitch -a codexWhat long-short-pitch does
The skill converts equity ideas into actionable long, short, or relative-value trade pitches. It defines where consensus is wrong, maps out catalyst timelines, and checks trade expression. The resulting framing includes sizing rules, add or trim triggers, short borrow dynamics, and concrete points that would invalidate the thesis.
Input gathering links to company filings, earnings call transcripts, internal research notes, and market data. Inventing numbers is strictly ruled out: any missing figure is tagged as an assumption or labeled TBD. Sizable deliverables export directly into a standalone HTML pitch report, while focused questions stay in conversation.
Formal investment committee memos are explicitly out of scope, as the repository routes finalized institutional write-ups to memo-builder.
What it helps with
- Frame long or short equity trade expressions for portfolio managers
- Stress-test an investment thesis against borrow fees, crowding, and weak catalysts
- Establish concrete add, trim, cover, and stop-out rules
- Generate standalone HTML trade pitch reports from fundamental sources
Example prompts for the agent
- Draft a short pitch for this stock focusing on borrow cost, crowding, and exit triggers
- Outline the variant perception and upcoming catalyst path for a long equity thesis
- Stress-test this pair trade and identify specific disconfirming data points
How to install long-short-pitch
1Skills CLI (Vercel)
npx skills add openai/plugins --skill long-short-pitch -a codex2Manually: put the files in place
.agents/skills/long-short-pitch/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add openai/plugins --skill long-short-pitch -a claude-code2Manually: put the files in place
~/.claude/skills/long-short-pitch/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/long-short-pitch/
Manually: put the files in place
.cursor/skills/long-short-pitch/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install openai/plugins long-short-pitch2Manually: put the files in place
.github/skills/long-short-pitch/SKILL.mdPlatform CLI
gemini skills install https://github.com/openai/plugins --path plugins/public-equity-investing/skills/long-short-pitchManually: put the files in place
.devin/skills/long-short-pitch/SKILL.mdThe legacy .windsurf/skills/ path still works
Claude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: Codex skills: how to install, where they live, how to call them
Questions about long-short-pitch
How do I install long-short-pitch in Codex?
Run npx skills add openai/plugins --skill long-short-pitch -a codex. The configuration installs to .agents/skills/long-short-pitch/SKILL.md.
When should I use long-short-pitch instead of memo-builder?
Use long-short-pitch when working out trade mechanics, variant perception, sizing, and exit discipline. Switch to memo-builder only when you need a formal investment committee document or client-facing memo.
How does the skill handle incomplete company data?
It never invents figures or outputs a blank template. Instead, it produces a provisional thesis, documents all assumptions, and highlights the exact missing datapoints required to raise conviction.
