agent-instructions
Use when writing project instructions for a coding agent (CLAUDE.md, AGENTS.md, or equivalent). Covers what belongs in them, what does not, structure, and…
Use when analyzing or constructing an options position. Covers the greeks in practical terms, implied volatility and its term structure, common structures and their true risk, and assignment and expiration mechanics.
$ npx -y skills add nimadorostkar/Claude-Skills-collection --skill options-strategy --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/options-strategyContext preview
The summary Claude sees to decide when to auto-load this skill.
Use when analyzing or constructing an options position. Covers the greeks in practical terms, implied volatility and its term structure, common structures and their true risk, and assignment and expiration mechanics.
name: options-strategy description: Use when analyzing or constructing an options position. Covers the greeks in practical terms, implied volatility and its term structure, common structures and their true risk, and assignment and expiration mechanics. metadata: category: finance version: 1.0.0 tags: [options, greeks, volatility, spreads, risk]
Construct and analyze options positions with an accurate view of what is actually being risked. Options positions frequently have a risk profile very different from the one the buyer had in mind, and the difference shows up at expiration.
1. **State the view precisely** — Direction, magnitude, *and* time frame. An option requires all three. "I think it goes up" is not sufficient to choose a structure; being right about direction and wrong about timing loses money. 2. **Check where implied volatility sits** — Buying options when IV is at the 90th percentile of its own history means you need a large move just to overcome the volatility crush. Selling when IV is at the 10th percentile means being paid very little for real risk. 3. **Choose the structure that matches** — Long options for a large, fast move with defined risk. Spreads to reduce cost and cap gain. Selling premium only when IV is elevated and the risk is genuinely defined. 4. **Compute the actual maximum loss** — Not the one you imagine. For a naked short option, it is unbounded, and that is not a figure of speech. 5. **Plan the exit before entry** — Including what happens at expiration. Positions held into expiration have assignment and pin risk that did not exist the day before. 6. **Check the liquidity of the contracts** — A liquid underlying can have illiquid options. A wide bid-ask spread is a cost you pay twice.
**The IV crush that catches everyone:**
Situation: earnings in 2 days. You expect a large move. Stock at $100. The 30-day at-the-money straddle costs $8.00. Implied volatility: 68% (its 52-week range is 24%-71%). The market has priced a ±8% move. For the long straddle to profit, the stock must move MORE than 8%. After the announcement, IV collapses to ~30% regardless of the direction of the move — the uncertainty that justified 68% has been resolved. Outcome if the stock moves 6% (a large move, and you were directionally right): Intrinsic value gained : ~$6.00 Value lost to IV crush : ~$7.20 (vega ~0.18 x 38 vol points) Net : approximately -$1.20 per share. A loss. You predicted the direction, the stock moved substantially, and you lost money. This is the standard outcome of buying options into earnings and it surprises people every quarter. The structure that expresses "large move" without paying for elevated IV is a calendar or a diagonal — long the far-dated option (less vega-sensitive to the front-month crush), short the front. Or simply take a stock position.
**Computing the real risk, not the imagined one:**
def position_risk(legs: list[OptionLeg], underlying: float) -> RiskProfile:
prices = np.linspace(underlying * 0.5, underlying * 1.5, 400)
pnl = np.array([sum(leg.payoff_at_expiry(p) for leg in legs) for p in prices])
has_naked_short_call = any(
leg.is_short and leg.is_call and not _is_covered(leg, legs) for leg in legs
)
return RiskProfile(
max_gain=float(pnl.max()),
# This is the number people get wrong.
max_loss=float("-inf") if has_naked_short_call else float(pnl.min()),
breakevens=[float(p) for p in _zero_crossings(prices, pnl)],
net_debit=sum(leg.cost for leg in legs),
# Aggregate greeks: the position's actual exposure, not each leg's.
net_delta=sum(leg.delta * leg.quantity * 100 for leg in legs),
net_theta=sum(leg.theta * leg.quantity * 100 for leg in legs),
net_vega=sum(leg.vega * leg.quantity * 100 for leg in legs),
warning=(A curated library of 137 production-grade skills for Claude and other AI coding agents. Every skill follows one structure, speaks with one voice, and earns its place by changing what the agent does.
Repo: nimadorostkar/Claude-Skills-collection
Use when writing project instructions for a coding agent (CLAUDE.md, AGENTS.md, or equivalent). Covers what belongs in them, what does not, structure, and…
Use when an agent needs state that survives a session or a context compaction. Covers what to persist, file-based memory, structuring notes for retrieval, and…
Use when automating agent behavior with lifecycle hooks. Covers hook events, deterministic enforcement of rules the model should not be trusted to remember,…
Use when packaging skills, commands, hooks, and MCP servers into a distributable plugin. Covers manifest structure, bundling, versioning, testing, and…
Use when writing a new skill for an AI agent. Covers scoping, description writing for reliable triggering, progressive disclosure, and the difference between a…
Use when reviewing or improving an existing agent skill. Covers triggering accuracy, content quality, redundancy with the base model, and measuring whether the…