@v587d/capital-generation 2.7.1 → 2.7.3

package/selected-skills/trading-skills/skills/position-sizing/SKILL.md

@@ -0,0 +1,83 @@
+---
+name: position-sizing
+description: Use when the user needs a conservative position size from account equity, risk budget, entry, stop, and trading friction before entering a trade.
+---
+
+# Position Sizing
+
+Use this skill before entering a trade when you need a defensible size instead of a gut-feel size.
+
+This skill will not:
+
+- tell the user whether the trade thesis is good
+- override liquidity, gap, or event-risk judgment with a formula
+- turn an aggressive stop or oversized conviction into a safe trade
+
+## Role
+
+Act like a conservative risk manager. Survival comes before conviction.
+
+## When to use it
+
+Use it when the user has a trade idea and needs to know:
+
+- how many shares, units, or contracts fit the risk budget
+- how slippage, fees, or contract multipliers change the math
+- whether the planned stop makes the size impractical
+- whether the trade is arithmetically small enough before asking whether it is strategically worth taking
+
+## Inputs and context
+
+Ask for:
+
+- account size or equity
+- max risk as percent or cash amount
+- entry price
+- stop price
+- instrument type if it affects contract value
+- optional slippage, commission, or "extra buffer" assumptions
+
+If any key input is missing, state what is missing and stop rather than invent it.
+
+## Analysis process
+
+1. Compute the per-unit risk from entry to stop.
+2. Add user-provided friction assumptions if they materially increase realized risk.
+3. Compute the maximum position size that stays inside the risk budget.
+4. Report the rounded-down size, estimated total risk, and any caveats.
+5. Flag cases where the stop is too tight, too wide, or structurally unclear.
+
+Use [references/methodology.md](references/methodology.md) for sizing conventions and caveats. Use [assets/trade-plan-template.md](assets/trade-plan-template.md) when the user wants a reusable planning format.
+
+For agents that support code execution, use [references/calculation-helpers.md](references/calculation-helpers.md) for the shared helper functions that cover fixed-fractional sizing, volatility-adjusted sizing, Kelly fraction context, and futures contract math.
+
+## Output structure
+
+Prefer this output order:
+
+1. `Inputs Used`
+2. `Sizing Method`
+3. `Sizing Math`
+4. `Position Recommendation`
+5. `Risk Caveats`
+
+Always include:
+
+- position size
+- total dollar risk
+- percent of account at risk
+- assumptions used
+- caveats about slippage, gaps, leverage, or contract multipliers
+
+Do not imply the size is "safe" just because it fits the arithmetic.
+
+## Best practices
+
+- do not promise that a mathematically valid size is strategically appropriate
+- do not ignore contract multipliers or event risk when they materially change exposure
+- do not replace the need for liquidation planning in fast or illiquid markets
+
+## Usage examples
+
+- "Use `position-sizing` for a $150,000 account risking 0.4% on a long entry at 84.20 with a stop at 81.90 and 0.10 slippage."
+- "Use `position-sizing` on ES futures with account size $80,000, max loss $500, long entry 5210.25, and stop 5199.75."

package/selected-skills/trading-skills/skills/position-sizing/assets/trade-plan-template.md

@@ -0,0 +1,15 @@
+# Trade Plan Template
+
+- Instrument:
+- Direction:
+- Account size:
+- Max risk (% or cash):
+- Entry:
+- Stop:
+- Profit target:
+- Expected slippage:
+- Commission or fees:
+- Event risk notes:
+- Final size:
+- Total planned risk:
+- Reasons to reduce size further:

package/selected-skills/trading-skills/skills/position-sizing/references/calculation-helpers.md

@@ -0,0 +1,13 @@
+# Calculation Helpers
+
+For code-execution environments, reuse the shared repo helper at `skills/_lib/calculations.py`.
+
+Relevant functions for this skill:
+
+- `fixed_fractional_position_size`
+- `volatility_adjusted_position_size`
+- `kelly_fraction`
+- `futures_contract_risk`
+- `price_risk_per_unit`
+
+Keep the skill output grounded in the user's stated risk budget, friction assumptions, and contract details rather than defaulting missing values.

package/selected-skills/trading-skills/skills/position-sizing/references/methodology.md

@@ -0,0 +1,28 @@
+# Position Sizing Methodology
+
+## Base formula
+
+The conservative core formula is:
+
+`position size = allowed trade risk / per-unit realized risk`
+
+Where:
+
+- allowed trade risk is either a cash number or a percentage of equity converted to cash
+- per-unit realized risk is not just entry minus stop if trading friction is meaningful
+
+## Friction matters
+
+When it is realistic, add:
+
+- expected slippage on entry and exit
+- spread cost for illiquid instruments
+- commission if it is large relative to the stop distance
+- a gap-risk buffer when the stop may not fill at the intended level
+
+## Practical cautions
+
+- Tight stops can create oversized positions that are fragile in real execution.
+- Wide stops can produce tiny positions that may not justify the trade.
+- Futures and options require multiplier awareness. Do not reuse equity math blindly.
+- A mathematically valid size can still be inappropriate if event risk is elevated.

package/selected-skills/trading-skills/skills/post-trade-debrief/SKILL.md

@@ -0,0 +1,99 @@
+---
+name: post-trade-debrief
+description: Orchestrate a disciplined post-trade workflow by reconstructing the original plan, reviewing execution and rule adherence, and deciding whether the lesson is trade-specific or part of a larger repeatable pattern.
+---
+
+# Post Trade Debrief
+
+Use this workflow skill when a trade has closed and the user wants one clear learning process instead of manually deciding whether to do a single-trade review, a pattern review, or both.
+
+This workflow will not:
+
+- grade the trade only by PnL
+- skip the original plan reconstruction just because the user remembers the outcome vividly
+- force pattern analysis when the sample is still too small
+
+## Role
+
+Act like a post-trade learning gatekeeper. Your job is to reconstruct what happened, identify the real lesson, and decide whether the issue belongs in a single-trade debrief or a broader journal pattern review.
+
+## When to use it
+
+Use it when the user wants to:
+
+- review a closed trade end to end
+- convert one trade outcome into a concrete lesson
+- decide whether a mistake is isolated or part of a recurring pattern
+- route the result into journal analysis only when the sample supports it
+
+## Inputs and context
+
+Ask for:
+
+- instrument and direction
+- original thesis, entry, stop, target, and size
+- actual entry, exit, and result
+- whether rules were followed
+- what happened around the trade, including catalyst or regime context
+- whether the user suspects this trade reflects a recurring issue
+
+Helpful but optional:
+
+- prior review notes
+- whether similar trades exist in the journal
+- setup tags, timeframe tags, or catalyst tags
+
+Use the user's materials first.
+
+If the original plan is missing, say that clearly and keep the debrief provisional rather than inventing it.
+
+## Workflow routing
+
+Use the smallest useful chain:
+
+1. Run `post-trade-review` on the closed trade.
+2. If the review exposes a recurring-looking mistake and the user has enough history, run `journal-pattern-analyzer`.
+3. If the review shows a structural flaw in the original setup, run `risk-reward-sanity-check` on the original structure.
+4. If the review shows oversizing or poor risk budgeting, run `position-sizing` on what the size should have been.
+
+Stop the workflow once the clearest lesson and next process change are established.
+
+## Decision logic
+
+Classify the result as:
+
+- `single-trade lesson`: the issue is important, but it belongs to this trade only for now
+- `pattern worth tracking`: the issue may recur and should be monitored in future reviews
+- `pattern confirmed`: the issue is strong enough to justify immediate journal-level rule changes
+
+## Output structure
+
+Prefer this output order:
+
+1. `Debrief Verdict`
+2. `Checks Run`
+3. `What Happened`
+4. `Main Lesson`
+5. `Is This A Pattern`
+6. `Process Change`
+7. `Next Skill Or Action`
+
+Always include:
+
+- the minimum set of checks actually used
+- the clearest lesson from the trade
+- whether the issue is isolated or part of a broader pattern
+- the next process change to test
+- whether the user should stop at this debrief or escalate to journal-level analysis
+
+## Best practices
+
+- do not let the workflow jump straight to pattern claims from one trade
+- do not skip reconstructing the original plan
+- do not turn the debrief into a motivational speech
+- do not recommend more than one or two process changes at once
+
+## Usage examples
+
+- "Use `post-trade-debrief` on this closed swing trade and tell me the real lesson."
+- "Use `post-trade-debrief` on this loss and tell me whether it is just one mistake or part of a pattern I need to address."

package/selected-skills/trading-skills/skills/post-trade-debrief/references/workflow-framework.md

@@ -0,0 +1,51 @@
+# Post-Trade Debrief Framework
+
+Use this framework to decide how far a trade review should go after a position is closed.
+
+## Core question
+
+The question is not only:
+
+- "Did I make or lose money?"
+
+The real questions are:
+
+- What was the original plan?
+- What actually happened?
+- What is the clearest lesson?
+- Is this lesson trade-specific or part of a larger pattern?
+
+## Typical routing
+
+Start with:
+
+- `post-trade-review`
+
+Add as needed:
+
+- `journal-pattern-analyzer` when the same type of issue appears often enough to justify pattern review
+- `risk-reward-sanity-check` when the original setup structure itself looks flawed
+- `position-sizing` when the main issue was oversized risk
+
+## Stop conditions
+
+Stop when:
+
+- the original plan cannot be reconstructed well enough for stronger claims
+- the sample is too small for pattern analysis
+- the clearest lesson is already specific and actionable at the single-trade level
+
+## Wrapper discipline
+
+The workflow should:
+
+- start with reconstruction, not outcome
+- separate single-trade learning from pattern learning
+- return one clear process change
+- avoid overstating what one trade proves
+
+## Boundaries
+
+This workflow does not reopen the trade.
+
+It only decides what should be learned from it.

package/selected-skills/trading-skills/skills/post-trade-review/SKILL.md

@@ -0,0 +1,82 @@
+---
+name: post-trade-review
+description: Guide a disciplined post-trade review across thesis quality, setup quality, execution, adherence, mistakes, and lessons without turning the result into hindsight theater.
+---
+
+# Post Trade Review
+
+Use this skill after a trade closes or after a meaningful trade sequence when you want an honest process review.
+
+The objective is to improve repeatability, not to rewrite history.
+
+This skill will not:
+
+- grade the trade only by PnL
+- excuse rule-breaking because the outcome was positive
+- turn hindsight into a fake lesson
+
+## Role
+
+Act like a process-focused trading coach. Reconstruct the plan, separate decision quality from outcome, and finish with concrete process improvements.
+
+## When to use it
+
+Use it when the user wants to understand:
+
+- whether the thesis was sound
+- whether execution matched the plan
+- whether size and rule adherence were acceptable
+- what mistake or strong decision mattered most
+
+## Inputs and context
+
+Ask for:
+
+- instrument and direction
+- original thesis
+- planned entry, stop, target, and size
+- actual entry, exit, and size
+- whether rules were followed
+- what happened around the trade
+- lesson the user currently believes they learned
+
+## Analysis process
+
+1. Reconstruct the original plan before judging the outcome.
+2. Separate thesis quality from execution quality.
+3. Identify rule adherence or violations.
+4. Classify mistakes as analytical, emotional, procedural, sizing, or market-environment related.
+5. End with one or two actionable process changes, not a motivational speech.
+
+Use [references/review-rubric.md](references/review-rubric.md) for the default rubric.
+
+## Output structure
+
+Prefer this output order:
+
+1. `Original Plan`
+2. `Execution Review`
+3. `Rule Adherence`
+4. `Main Mistake Or Best Decision`
+5. `Process Change`
+
+Always include:
+
+- original thesis summary
+- setup quality assessment
+- execution quality assessment
+- what was done as planned
+- what was not done as planned
+- biggest mistake or best process decision
+- one concrete lesson to carry forward
+
+## Best practices
+
+- do not grade the trade only by PnL
+- do not excuse broken process because the outcome happened to be positive
+- do not rewrite the original thesis with hindsight
+
+## Usage examples
+
+- "Use `post-trade-review` on my NVDA swing long where I sized correctly but moved the stop wider after entry."
+- "Use `post-trade-review` for a EURUSD short that hit target, but only after I chased the entry and doubled the original risk."

package/selected-skills/trading-skills/skills/post-trade-review/references/review-rubric.md

@@ -0,0 +1,34 @@
+# Review Rubric
+
+## Thesis
+
+- Was the original idea clear enough to invalidate?
+- Was the catalyst or setup actually relevant?
+
+## Execution
+
+- Did the entry match the plan?
+- Was the size appropriate for the setup quality and volatility?
+
+## Adherence
+
+- Was the stop honored?
+- Were targets managed according to the plan?
+- Were any rules changed mid-trade?
+
+## Mistakes
+
+Classify mistakes without drama:
+
+- process mistake
+- emotional override
+- poor sizing
+- weak thesis
+- wrong environment read
+
+## Lessons
+
+The best lesson is specific and repeatable:
+
+- "Reduce size by 25% into high-impact data days."
+- "Do not widen stops after entry unless the original plan explicitly allowed it."

package/selected-skills/trading-skills/skills/pre-trade-check/SKILL.md

@@ -0,0 +1,136 @@
+---
+name: pre-trade-check
+description: Orchestrate a disciplined pre-trade workflow by routing a watchlist or trade idea through the minimum set of underlying skills needed to decide whether the trade is ready, not ready, or should be resized or reworked first.
+---
+
+# Pre Trade Check
+
+Use this workflow skill when the user wants one disciplined answer to the question, "Is this trade actually ready?" rather than running each underlying check manually.
+
+This workflow will not:
+
+- place or stage orders
+- skip missing-information problems just to reach a yes or no answer
+- force every trade through every skill if the shorter path is sufficient
+
+## Role
+
+Act like a pre-trade gatekeeper. Your job is to run the smallest useful sequence of underlying skills, stop where the trade fails, and return a clear readiness verdict.
+
+## When to use it
+
+Use it when the user wants to:
+
+- check a trade idea end to end before entering
+- know whether the issue is the thesis, event risk, structure, execution, concentration, or size
+- turn a messy idea into a clear go / no-go / not-yet decision
+- avoid skipping important checks under time pressure
+
+## Inputs and context
+
+Ask for the minimum information needed to route the workflow:
+
+- whether the user has a single idea or a watchlist
+- instrument, direction, and intended timeframe
+- thesis or rough claim
+- entry, stop, and target if already known
+- any upcoming catalyst or event concern
+- account or portfolio context if concentration or size may matter
+
+Helpful but optional:
+
+- current regime observations
+- watchlist notes
+- order-type plan
+- constraints such as regular-hours only, no event holds, or maximum portfolio risk
+
+Use the user's materials first.
+
+If the user only has a rough idea, start with discovery and evidence checks instead of pretending the trade is already in construction mode.
+
+## Workflow routing
+
+Use the shortest chain that answers the user's actual problem:
+
+1. If the user starts with many names, run `watchlist-review`.
+2. If those names still need one event picture, run `catalyst-map`.
+3. If the idea is under-researched, run `evidence-gap-check`.
+4. If the claim is still not testable, run `thesis-validation`.
+5. If event risk matters, run `earnings-preview` or `macro-event-analysis`.
+6. If broader context matters, run `market-regime-analysis`.
+7. If structure is defined, run `risk-reward-sanity-check`.
+8. If order logic matters, run `execution-plan-check`.
+9. If portfolio overlap matters, run `portfolio-concentration`.
+10. If the trade is still viable and the stop is stable, run `position-sizing`.
+
+Stop the workflow as soon as a blocking issue appears and say why the trade is not ready yet.
+
+## Decision logic
+
+Classify the result as:
+
+- `ready`: the trade has passed the relevant checks and can move to entry discipline
+- `ready with conditions`: the trade may proceed only if the stated constraints are respected
+- `not ready yet`: the trade needs more research, cleaner structure, better execution logic, or smaller risk before it deserves action
+- `reject`: the current idea is weak enough that forcing it forward would be process failure
+
+## Output structure
+
+Prefer this output order:
+
+1. `Pre-Trade Verdict`
+2. `Checks Run`
+3. `Blocking Issue`
+4. `What Passed`
+5. `What Still Needs Work`
+6. `Updated Trade Context`
+7. `Next Skill Or Action`
+
+Always include:
+
+- the minimum set of checks actually used
+- the main reason the trade is ready or not ready
+- the most important unresolved issue
+- what the user should do next if the answer is not ready
+- a compact updated trade context block when enough information exists
+
+## Updated Trade Context
+
+When enough context exists, carry forward a compact block like this:
+
+```markdown
+## Trade Context
+
+- instrument:
+- direction:
+- timeframe:
+- thesis:
+- key_drivers:
+- key_risks:
+- upcoming_events:
+- invalidation:
+- entry_idea:
+- stop_idea:
+- target_idea:
+- position_size_hint:
+- portfolio_impact:
+- confidence_notes:
+- open_questions:
+- source_freshness:
+- assumptions:
+- next_recommended_skill:
+```
+
+Only populate the fields supported by the checks that actually ran. Do not invent missing fields just to make the block look complete.
+
+## Best practices
+
+- do not run every skill by default if two checks already answer the question
+- do not continue past a blocking issue without saying it is blocking
+- do not let a valid setup skip execution, concentration, or sizing when those are the real risks
+- do not let the wrapper hide which underlying skill produced the key concern
+
+## Usage examples
+
+- "Use `pre-trade-check` on this swing long idea in semis and tell me if it is actually ready."
+- "Use `pre-trade-check` on my watchlist for next week and route the best setup through the minimum checks needed before entry."

package/selected-skills/trading-skills/skills/pre-trade-check/references/workflow-framework.md

@@ -0,0 +1,61 @@
+# Pre-Trade Workflow Framework
+
+Use this framework to decide which checks a trade idea actually needs before entry.
+
+## Core principle
+
+Do not run a giant checklist blindly.
+
+Run the smallest set of checks that can answer:
+
+- Is the idea worth acting on?
+- If not, what is the blocking issue?
+
+## Typical routing
+
+If the user starts with many names:
+
+- `watchlist-review`
+- `catalyst-map`
+
+If the user has an interesting but underdeveloped idea:
+
+- `evidence-gap-check`
+- `thesis-validation`
+
+If event or regime risk matters:
+
+- `earnings-preview` or `macro-event-analysis`
+- `market-regime-analysis`
+
+If the user is already building the trade:
+
+- `risk-reward-sanity-check`
+- `execution-plan-check`
+- `portfolio-concentration`
+- `position-sizing`
+
+## Stop conditions
+
+Stop the workflow when:
+
+- the evidence is too incomplete
+- the thesis is too vague
+- the structure is incoherent
+- the order logic is not executable
+- the portfolio is already too concentrated
+
+## Wrapper discipline
+
+The wrapper should:
+
+- be explicit about which checks ran
+- stop at the real blocker
+- avoid fake certainty
+- produce one clear next action
+
+## Boundaries
+
+This workflow does not place trades.
+
+It only decides whether the idea deserves the next step.

package/selected-skills/trading-skills/skills/risk-reward-sanity-check/SKILL.md

@@ -0,0 +1,81 @@
+---
+name: risk-reward-sanity-check
+description: Use when the user wants to test whether a proposed entry, stop, and target structure is coherent, asymmetric enough, and vulnerable to obvious failure modes before the trade is placed.
+---
+
+# Risk Reward Sanity Check
+
+Use this skill when a trade idea exists but you want to inspect whether the structure makes sense before committing capital.
+
+This skill does not predict whether the trade will work. It checks whether the plan is internally coherent.
+
+This skill will not:
+
+- tell the user a positive ratio makes the trade good
+- replace thesis quality with a single numeric multiple
+- assume the stop or target is valid just because it is mathematically neat
+
+## Role
+
+Act like a skeptical pre-trade reviewer. Challenge the structure before money is at risk.
+
+## When to use it
+
+Use it when the user wants to know whether:
+
+- the stop and target make sense together
+- the proposed asymmetry is real or just cosmetic
+- the thesis actually supports the target
+- obvious structural issues are hiding behind attractive math
+
+## Inputs and context
+
+Ask for:
+
+- direction
+- entry
+- stop
+- one or more targets
+- thesis in one or two sentences
+- optional context such as time horizon, catalyst, or nearby event risk
+
+## Analysis process
+
+1. Calculate distance from entry to stop and from entry to target.
+2. Express the raw risk-reward ratio.
+3. Check for structural issues such as targets inside noise, stops placed at obvious liquidity pools, or targets with no thesis support.
+4. Explain what would invalidate the structure even if the ratio looks attractive on paper.
+5. Separate "good ratio" from "good trade." They are not the same thing.
+
+See [references/failure-modes.md](references/failure-modes.md) for the default checklist.
+
+The arithmetic here is small enough to show in the reply: distance from entry to stop, distance from entry to target, the ratio, and what slippage or a gap does to both. The analysis role in this plugin has no shell and runs no helper scripts, so when more than a handful of trades are involved, hand the brief (the rows, the columns, the statistic wanted) to the host's data side — in a role-split host, the agent that owns statistics — and use only the numbers that come back.
+
+## Output structure
+
+Prefer this output order:
+
+1. `Setup Summary`
+2. `Risk Reward Math`
+3. `Structural Review`
+4. `Decision Risk`
+5. `What Must Be True`
+
+Always include:
+
+- raw risk-reward math
+- asymmetry assessment
+- the main structural concern
+- what would need to be true for the target to be plausible
+- any mismatch between thesis, stop, and holding period
+
+## Best practices
+
+- do not tell the user a trade will win because the ratio looks attractive
+- do not replace thesis quality with a single numeric multiple
+- do not ignore catalyst or event risk that can break the planned structure
+
+## Usage examples
+
+- "Use `risk-reward-sanity-check` on a long entry at 58.20, stop 55.90, target 65.50, thesis: breakout continuation if software leadership holds."
+- "Use `risk-reward-sanity-check` for a short at 412.80 with stop 418.10 and two targets at 406.00 and 399.50 ahead of payrolls."