agentleFS
nimadorostkarnimadorostkar/Claude-Skills-collection/skills/finance/risk-management/SKILL.md

risk-management

Skill37 starsChanged 37 days agoMITView on GitHub

Use when defining or enforcing trading risk limits. Covers drawdown circuit breakers, exposure budgets, correlation risk, stop discipline, and pre-trade gates that prevent the worst decisions. Trigger on "set up risk rules for my account", "i'm in a drawdown, should i cut size", "when should i stop trading for the day". Do not use for sizing a single trade — see the position-sizing skill.

17 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use market-microstructure and sentiment-analysis.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • agiprolabsSkill

    market-microstructure

    DEX orderflow analysis, trade classification, buyer/seller pressure, and microstructure signals for Solana tokens

    Also in 3 projects that use this

  • MicrosoftSkill

    sentiment-analysis

    Use when a director of product asks for the fictional sentiment split and recent excerpts.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

3 rules. 0 are common, 3 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 15 edited versions we read keep these.

  • 3

    Distinctive

    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
  1. Distinctive:

    Before entering a trade, checking it against the rules.

    line 20
  2. Distinctive:

    After a loss, reviewing whether the rules were followed.

    line 21
  3. Distinctive:

    Stop-loss discipline and the failure modes around it.

    line 28

“Common” is measured: kept by at least 6 in 10 of the 15 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Risk Management

Purpose

Define the limits that keep a losing period from becoming a terminal one, and enforce them mechanically — because the moment they matter most is the moment you will most want to override them.

When to Use

  • Setting risk limits for a strategy or an account.
  • In a drawdown, deciding whether to reduce size.
  • Before entering a trade, checking it against the rules.
  • After a loss, reviewing whether the rules were followed.

Capabilities

  • Drawdown circuit breakers and de-risking schedules.
  • Exposure budgets: gross, net, per sector, per correlated group.
  • Correlation-aware risk aggregation.
  • Stop-loss discipline and the failure modes around it.
  • Pre-trade checklists as a hard gate.

Inputs

  • Account equity and the current drawdown from the high-water mark.
  • Open positions and their correlations.
  • The proposed trade.

Outputs

  • A pass/fail against each limit.
  • Required size reduction, if any.
  • A stated reason for a rejection.

Workflow

  1. Set the limits when calm — Maximum risk per trade, maximum total exposure, maximum drawdown before de-risking. Write them down. They must be decided before they are tested.
  2. Define the circuit breaker — A drawdown level at which size is cut, and a deeper level at which trading stops entirely. This is the rule that prevents a bad month from becoming a career-ending one.
  3. Aggregate correlated risk — Positions in the same sector, the same factor, or the same theme are one risk. Sum them.
  4. Gate every trade — Check it against the limits before entry, mechanically. A checklist that is consulted after the position is on is not a gate.
  5. Honor the stop — The stop is the risk. Moving it further away after entry converts a defined loss into an undefined one, and it is the single most common way accounts are destroyed.
  6. Review the rule, not the outcome — A trade that followed the rules and lost was a good trade. A trade that broke the rules and won was a bad one, and it will be repeated.

Best Practices

  • Widening a stop is the cardinal sin. The stop was set where the thesis is wrong; moving it means the thesis is wrong and you are staying anyway.
  • A drawdown circuit breaker must be mechanical. In a drawdown, judgment is impaired precisely when it is most needed — that is what the rule is for.
  • Reduce size in a drawdown, not after recovering from one. Risking the same percentage of a smaller account is already a reduction; cutting the percentage as well is what stops the compounding.
  • Correlation is not a static number. It rises sharply in a selloff, which is exactly when the diversification you were relying on stops existing.
  • The purpose of a risk limit is to be binding. A limit you override when it is inconvenient is not a limit; it is a preference.
  • Track rule adherence separately from profit and loss. They are different things, and only one of them is under your control.

Examples

A drawdown circuit breaker with a defined de-risking schedule:

@dataclass(frozen=True)
class RiskPolicy:
    base_risk_pct: float = 0.01           # 1% per trade at full size
    max_gross_exposure: float = 1.5       # 150% of equity, long + short
    max_sector_exposure: float = 0.30
    max_correlated_risk_pct: float = 0.04 # aggregate risk across correlated positions


def size_multiplier(drawdown_pct: float) -> float:
    """De-risk as the drawdown deepens. Mechanical, not discretionary."""
    if drawdown_pct >= 0.20:
        return 0.0        # STOP. No new positions. Review the strategy, not the market.
    if drawdown_pct >= 0.15:
        return 0.25
    if drawdown_pct >= 0.10:
        return 0.50
    if drawdown_pct >= 0.05:
        return 0.75
    return 1.0


# Equity 100,000, high-water 120,000 -> drawdown 16.7% -> multiplier 0.25.
# Risk per trade is now 0.25% of 100,000 = $250, not the $1,000 it was at the peak.
# This is a 4x reduction in position size at the point where the strategy is
# demonstrably not working. It is not pessimism; it is arithmetic.

A pre-trade gate that is actually a gate:

def pre_trade_check(trade: ProposedTrade, book: Portfolio, policy: RiskPolicy) -> GateResult:
    failures = []

    if trade.stop is None:
        failures.append("BLOCK: no stop defined. Risk is undefined and unbounded.")

    if trade.dollar_risk > book.equity * policy.base_risk_pct * size_multiplier(book.drawdown):
        failures.append(
            f"BLOCK: risk ${trade.dollar_risk:,.0f} exceeds the current limit "
            f"(drawdown {book.drawdown:.1%} -> size multiplier "
            f"{size_multiplier(book.drawdown):.2f})"
        )

    sector_after = book.sector_exposure(trade.sector) + trade.position_value / book.equity
    if sector_after > policy.max_sector_exposure:
        failures.append(
            f"BLOCK: {trade.sector} exposure would reach {sector_after:.0%} "
            f"(limit {policy.max_sector_exposure:.0%})"
        )

    # Correlated positions are one position.
    correlated = book.positions_correlated_with(trade.symbol, threshold=0.7)
    total_correlated_risk = sum(p.dollar_risk for p in correlated) + trade.dollar_risk
    if total_correlated_risk > book.equity * policy.max_correlated_risk_pct:
        failures.append(
            f"BLOCK: aggregate risk across {len(correlated) + 1} correlated positions "
            f"({', '.join(p.symbol for p in correlated)}) would reach "
            f"${total_correlated_risk:,.0f}"
        )

    if trade.is_earnings_within(days=2) and not trade.earnings_intentional:
        failures.append("BLOCK: earnings within 2 days. Overnight gap risk is not sized for.")

    return GateResult(passed=not failures, failures=failures)

Notes

  • The gate above blocks on a missing stop. That single check prevents the majority of catastrophic single-position losses, because an unbounded loss requires an undefined risk.
  • A 20% drawdown triggering a full stop is not a panic response; it is a recognition that either the strategy has stopped working or the market regime has changed, and both warrant investigation rather than more trades.
  • This is educational material on risk-management methodology, not financial advice. Limits appropriate for one account and strategy may be entirely wrong for another.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.