profit-structure
Skill2 starsChanged 17 days agoMITView on GitHub
Writes an equity analyst's first page on a listed company - where its operating profit comes from, line by line with dollar sizes and filing sources, and which lines a government decision can move (tax, tariff, subsidy, rate approval, licence, wage law, procurement) and who decides each (Congress, President, agency, governor, legislature, commission, attorney general, voters). Use when the user asks what drives a company's profit, how exposed it is to policy or regulation, or before an election-risk analysis.
In this repository5 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- candidate-positionsSkill
- election-riskSkill
- prediction-marketsSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 12
Start from how the company makes money, not from the news or the markets.
- Distinctive: line 38
Do not look at prediction markets or candidate positions while writing this.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Profit structure
Start from how the company makes money, not from the news or the markets. Rough is fine; this is an analyst's first page, not an audit. Do not invent numbers: a line you cannot size says "not disclosed".
erisk below means python3 <skill-dir>/scripts/erisk.py (${CLAUDE_SKILL_DIR} in Claude Code). Run it from
the user's working directory. Output: ./election-risk-out/<TICKER>/structure.md.
Workflow
- [ ] 1. Base: latest annual operating income (or adjusted EBIT / EBITDA when that is what the company
steers by; say which), with the filing URL
- [ ] 2. Where operating profit comes from: segments, products, geographies, big cost lines
- [ ] 3. Policy lines inside them: taxes, credits, subsidies, tariffs, regulated rates, licences, wage laws,
procurement, compliance costs
- [ ] 4. Which lines a government can move, the lever, who decides, and at what level
- [ ] 5. Write structure.md (format below)Sources. The latest 10-K (or 20-F/40-F, or the annual report for non-US companies) and the latest earnings
release. For US filers, erisk filings edgar <TICKER> gives headline numbers; read the segment note and risk
factors for the rest. How to find and read the numbers: references/filings.md.
Split by whatever matters for this company: US states where they are material (staff, stores, rate base,
sales), customer type (government, Medicaid, Medicare), countries of import. Every line: $ per year, source
URL, verified (from a filing) or estimated (your arithmetic, shown).
Do not look at prediction markets or candidate positions while writing this. A structure written with the markets open drifts toward whatever has a market.
structure.md format
# <Company> (<TICKER>): profit structure
Written <date>. Base: <FY> <measure> $<amount> (<status>, <url>)
## Where operating profit comes from
| # | Line | $ / yr | Source | Status |
## Policy lines inside those
| # | Line | $ / yr | Source | Status |
## What a government can move
| Line | Lever | Who decides | Level (federal / state: which states / local) | Issue tag |
## Locked in
Things already law that no election changes, with size.Issue tag uses the policy book's words (utilities, data-centers, energy-oil-gas, clean-cars-ev,
gaming-betting, health-medicaid-aca, insurance, labor-min-wage, gig-workers,
tech-antitrust-privacy, kids-online, crypto, film-tax-credits, corporate-sales-property-tax,
tariffs-trade, defense-procurement, housing, cannabis, transport-ai-vehicles,
wildfire-liability, other: <what>), so $candidate-positions and $election-risk can look lines up.
Answer format
The "What a government can move" table, biggest lines first, then one line: the base and the share of profit the movable lines touch. Offer the full file path.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.