trade-performance-coach

---
name: trade-performance-coach
description: Use when a trade has closed, or monthly, and the user wants a coaching review of process adherence, risk discipline, execution quality, or behavior patterns — never stock picks or trade approval.
category: Performance
version: 1.0.0
tools: []
---

# Trade Performance Coach

Review recorded trade outcomes and journal evidence to help a trader improve their decision process — filling the role a risk manager, desk lead, or trading coach would provide. This is strictly a process-review skill: it never recommends entering, exiting, buying, selling, shorting, holding, or sizing a specific security.

## When to Use

Use when: a trade has closed and the user wants a post-trade coaching review; a partial close occurred and the user wants to inspect sizing, stop, or exit behavior; the user has `trader-memory-core` thesis records and `signal-postmortem` findings and wants next-session operating rules; the user wants a monthly review of recurring process, risk, execution, or behavior patterns; the user wants a risk-manager-style review of their own recorded trades; the user asks whether a loss was a process error, execution error, market-environment issue, or acceptable variance; or the user wants possible FOMO, revenge-trade, overconfidence, hesitation, stop-moving, or size-creep patterns flagged with evidence.

## When Not to Use

Do not pick stocks or rank trade candidates, approve or reject a live trade as financial advice, place orders or draft broker instructions, provide therapy or mental-health diagnosis, infer private psychological traits beyond the evidence supplied, or shame the user for losses or rule violations. This skill doesn't replace `trader-memory-core` — it consumes journal/thesis records and produces coaching findings from them. If input is incomplete, default to a `REVIEW_REQUIRED` or journal-only mode and ask for missing records rather than inventing evidence.

## Inputs

Useful upstream records: a `trader-memory-core` closed thesis or journal entry, `signal-postmortem` findings, the original trade plan/ticket, the actual entry/exit/partial-close actions, the user's risk plan, and optionally market-regime context. Minimum useful input is one recorded trade or one monthly aggregate; the review tolerates partial records and marks missing evidence as "unclear" rather than guessing.

Preferred fields per trade: review type (single trade / partial close / monthly aggregate), trade ID, ticker, outcome (win/loss/breakeven/mixed); planned thesis, entry, stop, target, risk in R, whether the thesis was recorded before entry, whether the setup was confirmed, and the market regime at the time; actual entry, exit, risk in R, portfolio heat in R, whether the stop was moved (and whether that move was pre-planned), whether entry happened before confirmation, and whether the trade went against the regime; the user's risk plan (max risk per trade in R, max portfolio heat in R, max weekly loss in R); postmortem root cause (thesis quality / execution / risk sizing / market environment / rule violation / randomness / unknown); and journal reflection/emotions.

## Workflow

**Evaluate process adherence** against the user's documented plan and rules: missing pre-entry thesis, setup confirmation skipped, trade taken against the market-regime gate, stop moved without a pre-defined rule, exit/partial close inconsistent with plan, or incomplete record quality.

**Evaluate risk discipline** against the risk plan: per-trade risk above max, portfolio heat above max, weekly-loss or consecutive-loss escalation, oversized trade after a winner or loser, and correlated exposure if provided.

**Evaluate execution quality** by classifying entry, stop, exit, add, trim, and review behavior — separating clean-process losses from genuine execution mistakes.

**Detect possible behavior patterns** from journal notes and action flags, always tying each tag to specific evidence and using non-diagnostic ("possible pattern") language. Supported tags: `fomo_entry`, `revenge_trade`, `premature_exit`, `overconfidence_after_winner`, `stop_moved`, `size_creep`, `hesitation`, `rule_drift`, `no_pattern_detected`.

**Produce next-session operating rules** — temporary, concrete guardrails such as requiring a thesis record and screenshot before the next entry, capping risk at 0.5R for the next two trades after a rule violation, switching to review-only mode after repeated revenge-trade evidence, or adding a missed entry to the watchlist instead of chasing it.

**End with a human decision gate.** Default action is `journal_only`; other allowed actions are `accept_rules`, `modify_rules`, and `defer`.

## Output

Return, in chat, a report covering: schema version, review type, review ID, overall verdict, summary, scores, process-adherence findings, risk-manager notes, execution-quality assessment, behavioral pattern tags, next-session operating rules, coach questions for reflection, the human decision gate, and a disclaimer.

Verdicts:

| Verdict | Meaning |
|---|---|
| `OK` | No material process violation found; outcome compatible with the plan |
| `WARN` | Minor process or record-quality concern |
| `REVIEW_REQUIRED` | Meaningful process, risk, or behavior finding before a similar trade |
| `RULE_VIOLATION` | Explicit user rule appears to have been broken |
| `COOL_DOWN` | Repeated violations, drawdown/revenge pattern, or escalation suggests review-only mode |

## Guardrails

This is process-review support, not financial advice. Never recommend buying, selling, shorting, holding, or sizing a specific security. Never provide therapy or mental-health diagnosis, or infer personality traits. Never shame or moralize. Tie every behavior tag to evidence and use "possible pattern" language. Always include a human decision gate. Default to journal/review mode when data is incomplete.