@v587d/capital-generation 2.7.1 → 2.7.3

package/selected-skills/trading-skills/skills/risk-reward-sanity-check/references/calculation-helpers.md

@@ -0,0 +1,14 @@
+# Calculation Helpers
+
+What to work out, and where the numbers come from in this plugin: this much arithmetic is small
+enough to show in the reply, and the analysis role has no shell and runs no helper scripts. When more
+than one trade is involved, hand the rows and the statistic wanted to the host's data side — in a
+role-split host, the agent that owns statistics — and use only what comes back.
+
+The measures this skill cares about:
+
+- `risk_reward_ratio` — the distance from entry to target divided by the distance from entry to stop,
+  per unit of the instrument (a contract multiplier scales both sides alike)
+- `expectancy` — win rate times average win, minus the loss rate times average loss
+
+Use the math only as a cross-check for the proposed structure. Do not let a favorable ratio override weak thesis support, event risk, or unrealistic target logic.

package/selected-skills/trading-skills/skills/risk-reward-sanity-check/references/failure-modes.md

@@ -0,0 +1,9 @@
+# Common Failure Modes
+
+- Attractive ratio with no credible path to the target
+- Stop placed at an obvious level that is likely to be swept before thesis confirmation
+- Target based on hope rather than structure, catalyst, or historical travel
+- Stop so wide that the trade becomes low-conviction by definition
+- Time horizon mismatch: intraday stop with multi-week target or vice versa
+- Ignoring event risk that can break the intended structure
+- Multiple targets that were not tied to a position-management plan

package/selected-skills/trading-skills/skills/thesis-validation/SKILL.md

@@ -0,0 +1,123 @@
+---
+name: thesis-validation
+description: Pressure-test a trade or investment thesis by clarifying the core claim, evidence, invalidation, timeframe, and dependency chain before the user turns it into an entry, stop, or size.
+---
+
+# Thesis Validation
+
+Use this skill before trade construction when the user has an idea, narrative, or directional lean but needs to know whether the thesis is clear enough to trust.
+
+This skill will not:
+
+- replace missing evidence with confident prose
+- choose entry, stop, target, or size
+- turn a plausible narrative into a validated edge
+
+## Role
+
+Act like a skeptical thesis reviewer. Your job is to make the claim testable, separate evidence from assumption, and show what would break the idea.
+
+## When to use it
+
+Use it when the user wants to:
+
+- turn rough conviction into a falsifiable thesis
+- check whether the evidence actually supports the claim
+- identify the catalysts, dependencies, and weak links that matter
+- decide whether the idea is ready for `risk-reward-sanity-check`, `position-sizing`, or more research first
+
+## Inputs and context
+
+Ask for:
+
+- instrument, theme, or market being discussed
+- thesis in one to three sentences
+- intended timeframe
+- why now, including any catalysts, regime assumptions, or timing logic
+- the strongest evidence supporting the thesis
+- what the user currently believes would invalidate it
+
+Helpful but optional:
+
+- supporting notes, transcripts, charts, watchlist context, or prior research
+- known counterarguments
+- whether the user is thinking like a trader, investor, or both
+
+Use the user's materials first.
+
+If the core claim, timeframe, or invalidation logic is missing, say exactly what is missing and keep the validation provisional rather than inventing it.
+
+Test the claim against what the user put in front of you. When a fact would genuinely change the verdict, do not go looking for it from here: write the exact brief (the claim, the missing field, the as-of date) and hand it to the host's own retrieval path — in a role-split host, the agent that owns retrieval — and keep the validation provisional for whatever does not come back.
+
+## Analysis process
+
+1. Restate the thesis as a clear claim, not a vibe or slogan.
+2. Separate observed facts, inference, and assumption.
+3. Check whether the stated timeframe matches the evidence and catalysts.
+4. Identify the dependencies that must remain true, such as macro backdrop, leadership, earnings follow-through, valuation tolerance, or liquidity conditions.
+5. State what evidence would invalidate the thesis, not just what would create temporary noise.
+6. Surface the strongest counterargument and the biggest missing evidence gap.
+7. Conclude whether the idea is ready for trade construction, needs narrower framing, or needs more research first.
+
+Use [references/thesis-checklist.md](references/thesis-checklist.md) when you need a compact checklist for claim quality, timeframe fit, and invalidation discipline.
+
+## Core Assessment Framework
+
+Assess the thesis on five anchors before calling it actionable:
+
+- `Claim Clarity`: can the thesis be stated as one falsifiable claim. Example: "AI capex strength should support semis leadership over the next six weeks" is clearer than "AI is bullish."
+- `Evidence Quality`: is the support specific, relevant, and current for the chosen timeframe. Example: a recent earnings guide and sector breadth evidence is stronger than a months-old headline.
+- `Timeframe Fit`: does the holding period match the evidence and catalyst cadence. Example: a one-day trade justified by a twelve-month fundamental story is a mismatch.
+- `Dependency Load`: how many external conditions must stay true. Example: an idea that requires lower yields, broad risk appetite, and flawless earnings read-through has a heavier dependency load than a single-company catalyst thesis.
+- `Invalidation Quality`: is there a concrete condition that would weaken or break the claim. Example: "supplier orders slow and leadership breaks after earnings" is stronger than "it just feels wrong."
+
+Use the anchors to classify:
+
+- `ready for structure review`: the claim is clear, evidence is relevant, timeframe fits, and invalidation is concrete enough to move into trade construction
+- `promising but incomplete`: the idea may be interesting, but a key evidence gap, timeframe mismatch, or weak invalidation rule needs fixing first
+- `weak thesis`: the claim is too vague, too assumption-heavy, or too dependent on untested conditions to support a disciplined trade or investment plan
+
+## Evidence That Would Invalidate This Analysis
+
+- the user's actual timeframe changes enough to make the current evidence irrelevant
+- a key catalyst date, dependency, or macro assumption changes materially
+- the thesis turns out to rely on missing evidence that was not actually present in the shared context
+- new information strengthens the strongest counterargument more than the original thesis
+- the user reframes the idea from tactical trade to long-horizon investment, or the reverse
+
+## Output structure
+
+Prefer this output order:
+
+1. `Validation Summary`
+2. `Core Claim`
+3. `Evidence For`
+4. `Evidence Against`
+5. `Invalidation And Timeframe`
+6. `Dependencies And Catalysts`
+7. `Open Questions`
+8. `Next Skill`
+
+Always include:
+
+- the cleaned-up thesis in plain language
+- the strongest supporting evidence
+- the strongest weakening evidence or counterargument
+- what would invalidate the thesis
+- whether the timeframe fits the evidence
+- the most important dependency or catalyst
+- whether the idea should move next to `market-regime-analysis`, `risk-reward-sanity-check`, `position-sizing`, or more research
+
+A skill named above is an option for the user, not a step this role can take: only the skills the user switched on exist in this session, and the ones this plugin deliberately does not offer — placing orders, setting position size — are not reachable at all. Say what you would check next and let the user decide.
+
+## Best practices
+
+- do not let conviction substitute for evidence
+- do not confuse a catalyst with proof that the thesis is correct
+- do not rewrite the user's claim into something smarter than they actually argued
+- do not spill into entry, stop, target, or sizing decisions
+
+## Usage examples
+
+- "Use `thesis-validation` on this swing idea: semis leadership should continue for the next four to six weeks if AI demand stays firm and yields do not spike."
+- "Use `thesis-validation` on my 12-month investment thesis for ASML and tell me what evidence is real, what is assumption, and what would invalidate it."

package/selected-skills/trading-skills/skills/thesis-validation/references/thesis-checklist.md

@@ -0,0 +1,56 @@
+# Thesis Checklist
+
+Use this checklist to keep a thesis specific, testable, and reusable.
+
+## Core questions
+
+- What exactly is the claim?
+- Why should that claim be true?
+- Why now?
+- What timeframe does the claim belong to?
+- What would weaken it?
+- What would break it?
+
+## Evidence discipline
+
+Separate three buckets clearly:
+
+- `facts`: what is actually observed in the user's notes, filings, earnings, tape, or macro context
+- `inference`: what those facts may imply
+- `assumptions`: what still has to be true for the thesis to work
+
+If the thesis depends mostly on assumptions, it is not ready for trade construction.
+
+## Timeframe fit
+
+- Short-horizon trade theses should lean on near-term catalysts, tape behavior, regime, and timing.
+- Medium-horizon swing theses should connect the idea to a realistic catalyst path and a manageable dependency set.
+- Long-horizon investment theses should not be validated using only short-term price action.
+
+## Invalidation quality
+
+Good invalidation is concrete:
+
+- a thesis-critical data point fails
+- a key catalyst delivers the wrong read-through
+- leadership, breadth, or macro conditions change enough to break the logic
+- management guidance, demand, margins, or policy backdrop invalidates the core claim
+
+Bad invalidation is vague:
+
+- "the market disagrees"
+- "I lose confidence"
+- "it stops working"
+
+## Boundaries
+
+This checklist is for thesis quality only.
+
+It does not answer:
+
+- where to enter
+- where to place a stop
+- how large to trade
+- whether the portfolio can absorb the risk
+
+Those belong to downstream skills.

package/selected-skills/trading-skills/skills/watchlist-review/SKILL.md

@@ -0,0 +1,119 @@
+---
+name: watchlist-review
+description: Review a watchlist and rank which names deserve active attention, background monitoring, or removal based on catalysts, tradability, redundancy, and evidence quality for the user's style and timeframe.
+---
+
+# Watchlist Review
+
+Use this skill when the user already has a list of names, sectors, or themes and needs to narrow it into a smaller set of names worth preparing for, not just collecting.
+
+This skill will not:
+
+- predict which stock will outperform
+- replace deep single-name research or a full investment memo
+- turn a large list of tickers into a recommendation to trade them all
+
+## Role
+
+Act like a disciplined watchlist editor. Your job is to reduce noise, surface the names that actually matter, and explain why other names should stay in the background or come off the list.
+
+## When to use it
+
+Use it when the user wants to:
+
+- rank a watchlist by actionability instead of headline familiarity
+- cut a bloated list down to a smaller active set
+- identify which names deserve deeper work before the next session, week, or earnings cycle
+- separate high-interest names from duplicate, illiquid, or weakly supported ideas
+
+## Inputs and context
+
+Ask for:
+
+- the watchlist itself: tickers, companies, sectors, or themes
+- the user's style and timeframe: day trade, swing, event-driven, long-term investing, and so on
+- what the user is looking for: breakout candidates, earnings setups, valuation ideas, defensive rotation, macro sensitivity, and so on
+- any catalysts or dates already known
+- any liquidity or instrument constraints
+
+Helpful but optional:
+
+- notes on thesis quality or current levels
+- whether the watchlist is for idea discovery, active trade prep, or long-term monitoring
+- names the user already suspects are redundant
+
+Use the user's materials first.
+
+If the user provides only a vague theme and no list or criteria, say what is missing and keep the review limited rather than pretending to screen the entire market from scratch.
+
+Rank from the list the user handed you. When a name needs a live quote, a filing date or a freshness check, do not pull it from here: write the exact brief (ticker, the field needed, the as-of date) and hand it to the host's own retrieval path — in a role-split host, the agent that owns retrieval — and treat anything that does not come back as missing rather than assumed.
+
+## Analysis process
+
+1. Reconstruct the watchlist and the user's objective.
+2. Group names by sector, theme, catalyst, or market role.
+3. Identify which names have the clearest reason to stay on the active list for the user's timeframe.
+4. Demote names that are redundant, low-liquidity, weakly supported, or lacking a relevant catalyst.
+5. Separate names that need immediate prep from names that only need background monitoring.
+6. Explain what extra work each top-priority name still needs before trade construction or deeper research.
+7. End with a smaller, higher-signal watchlist and the recommended next skill for each top name.
+
+Use [references/review-framework.md](references/review-framework.md) when you need the default checklist for catalysts, tradability, redundancy, and preparation depth.
+
+## Core Assessment Framework
+
+Assess each candidate on five anchors before prioritizing it:
+
+- `Catalyst Relevance`: whether there is a concrete reason this name matters now for the user's timeframe. Example: upcoming earnings, sector leadership, policy sensitivity, or a thesis milestone is stronger than generic familiarity.
+- `Tradability And Fit`: whether the instrument suits the user's style, liquidity needs, and execution tolerance. Example: a thin small-cap may not belong on a day-trader's active list even if the story is interesting.
+- `Setup Or Thesis Readiness`: whether the user has enough evidence to justify further work. Example: a clear setup or investment claim deserves more attention than a name included only because it is "popular."
+- `Redundancy`: whether the name adds something distinct or just duplicates an existing holding, ETF exposure, or another watchlist name. Example: five semiconductor names with the same catalyst chain do not all deserve top billing.
+- `Preparation Depth`: whether the user already has the notes, catalyst dates, and open questions needed to act. Example: a name with known dates and a defined question is more actionable than one with no prep.
+
+Use the anchors to classify:
+
+- `active focus`: deserves near-term prep, deeper analysis, or trade planning
+- `monitor`: worth keeping in the background, but not yet important enough for active prep
+- `remove or defer`: currently too redundant, too vague, too illiquid, or too unsupported to earn attention
+
+## Evidence That Would Invalidate This Analysis
+
+- the user's timeframe or objective changes materially
+- a catalyst appears, disappears, or moves enough to change priority
+- liquidity, regime, or event conditions change enough to alter tradability
+- the watchlist turns out to be only part of the real decision set
+- new information makes a previously redundant or weak name meaningfully distinct
+
+## Output structure
+
+Prefer this output order:
+
+1. `Watchlist Summary`
+2. `Active Focus`
+3. `Monitor`
+4. `Remove Or Defer`
+5. `Preparation Gaps`
+6. `Next Skill`
+
+Always include:
+
+- the user's objective and timeframe
+- which names deserve active attention now
+- why the top names matter
+- which names are redundant, low-priority, or not actionable yet
+- what additional prep each top name needs
+- whether the next step is `earnings-preview`, `macro-event-analysis`, `market-regime-analysis`, `thesis-validation`, or no further action yet
+
+A skill named above is an option for the user, not a step this role can take: only the skills the user switched on exist in this session, and the ones this plugin deliberately does not offer — placing orders, setting position size — are not reachable at all. Say what you would check next and let the user decide.
+
+## Best practices
+
+- do not confuse a long watchlist with real opportunity
+- do not let familiar large-cap names crowd out clearer setups
+- do not keep multiple names on the active list if they are effectively the same exposure
+- do not keep names on the active list without a reason they matter now
+
+## Usage examples
+
+- "Use `watchlist-review` on these names for next week: NVDA, AMD, AVGO, SMCI, ASML, and TSM. I care about swing opportunities and earnings read-through."
+- "Use `watchlist-review` on my long-term investing watchlist and tell me which names actually deserve deeper thesis work this month."

package/selected-skills/trading-skills/skills/watchlist-review/references/review-framework.md

@@ -0,0 +1,57 @@
+# Watchlist Review Framework
+
+Use this framework to turn a long list of names into a smaller list that actually deserves attention.
+
+## What a good watchlist does
+
+A good watchlist is not a scrapbook.
+
+It should help the user answer:
+
+- Which names matter now?
+- Why do they matter now?
+- What would I do with more research on each name?
+- Which names are only background context?
+
+## Core triage questions
+
+For each name, ask:
+
+- Is there a catalyst, regime condition, or thesis milestone that makes this relevant now?
+- Is it tradable or investable for the user's style?
+- Does it add distinct information, or is it duplicative?
+- Is there enough preparation already done to justify deeper work?
+- What skill should come next if the name stays on the active list?
+
+## Typical reasons to keep a name active
+
+- upcoming earnings or other scheduled catalyst
+- sector leadership or weakness that matters to the user's book
+- distinct exposure the user does not already have elsewhere
+- a clear thesis or setup that is close to actionable
+- macro sensitivity that makes the name important for the next session or week
+
+## Typical reasons to demote or remove a name
+
+- no clear "why now"
+- duplicates another name or ETF already on the list
+- too illiquid for the user's style
+- no working thesis, no catalyst, and no next step
+- too much work remaining for the current timeframe
+
+## Active vs monitor vs remove
+
+- `active focus`: the user should do more prep soon because the name could matter in the stated timeframe
+- `monitor`: useful context, but not worth immediate deep work
+- `remove or defer`: not actionable enough to justify attention right now
+
+## Boundaries
+
+This framework does not decide:
+
+- whether the trade will work
+- what order type to use
+- how large the position should be
+- whether the portfolio can absorb the risk
+
+It only decides which names deserve scarce attention.