Copy a prompt. Your agent does the rest.

Use your TradeLoop tools to get [SYMBOL]'s current price, today's % change, and today's volume vs its recent average. Answer in one line, and note the data timestamp.

Use your TradeLoop tools to pull today's % change and volume for my watchlist: [SYMBOLS]. Rank by absolute move, flag anything with unusual volume, and note the data timestamp. Output a compact table.

Use your TradeLoop tools to scan today's market movers with the large-cap filter (price ≥ $20, dollar volume ≥ $100M) — top 10 gainers and top 10 losers. Output two short tables with % change and dollar volume, and note whether the data is live or end-of-day.

Use your TradeLoop tools to run a technical read on [SYMBOL] (daily timeframe): trend vs the 50/200 SMAs, RSI, MACD, and the overall confluence score with its confidence. Give me 3 bullets max — direction, momentum, and the one caveat that matters most. Research only, no buy/sell call.

Use your TradeLoop tools to get [SYMBOL]'s key levels: current price with timestamp, prior-day high/low/close, today's classic pivot points (PP, R1/S1), and the 52-week high/low with % distance from each. Output a short table.

Use your TradeLoop tools to find [SYMBOL]'s next earnings date, the consensus EPS estimate, and how the last 4 quarters landed vs estimates. Two lines: the date (and whether it's before/after the bell if known), then the beat/miss pattern.

Use your TradeLoop tools to pull the latest news for [SYMBOL] (last 24-48h). Give me the 5 most substantive headlines — skip listicles and recycled takes — each with source, timestamp, and a one-line summary. Label which are company statements vs press interpretation.

Use your TradeLoop tools to check BTC perp positioning: current funding rate (and what it means — who pays whom), open interest with its 24h/7d trend, and the recent liquidation skew. Three lines, one per signal, plain English. Research signal only.

Use your TradeLoop tools to get current prices, 24h change, and 24h volume for [COINS]. One table, sorted by market cap.

Use your TradeLoop tools to pull the latest readings for: the 10-year Treasury yield, the 10y-2y curve spread, and VIX. For each, give the latest value, the direction over the past month, and one clause on what it implies for risk appetite. Fact vs interpretation clearly separated.

Use your TradeLoop tools to list [SYMBOL]'s 3 most recent SEC filings — form type, filing date, and document link. If the newest is an 8-K, tell me which items it reports and what that item code means.

Use your TradeLoop tools to render a daily candlestick chart for [SYMBOL] with SMAs, Bollinger Bands, RSI, and MACD, then tell me in 2 lines what the visual structure shows (trend, and where price sits vs the bands).

Use your TradeLoop tools to take an order-flow snapshot of [COIN]: book imbalance (top levels), trade-sign imbalance from the recent tape, and the spread. One line per signal plus a one-line combined read (book and tape agree / disagree). Short-horizon research signal only.

Use your TradeLoop tools to get [SYMBOL]'s dividend picture: current yield, the next ex-dividend and payment dates, and the last few payments (is the dividend growing?). One short block; note the data timestamps.

Use your TradeLoop tools (edgar insider_trades) to pull [SYMBOL]'s recent insider transactions: open-market buys vs sells with sizes and who traded. Summarize the net direction — facts only, and separate grants/awards from open-market activity.

Use your TradeLoop tools to show [ETF]'s top-10 holdings with weights and its sector breakdown. Flag how concentrated the top 10 is as a % of the fund.

Use your TradeLoop sentiment tools to give me today's market mood: the equity Fear & Greed score and which of its 7 components are driving it, the crypto Fear & Greed, and the VIX read (level, percentile, term structure). One short paragraph — is sentiment at an actionable extreme or mid-range noise? Research context only, no trade calls.

Use your TradeLoop fred tools to pull net_liquidity. Tell me: the current level, whether it expanded or contracted over the last month and quarter, and which component (balance sheet, TGA, RRP) moved it. One paragraph, plain english, with the caveat that this is loose macro context.

Use your TradeLoop options tools to compute max pain for [SYMBOL] (nearest expiration): the max-pain strike vs current price, the call wall and put wall, and how concentrated open interest is. Two sentences + the key strikes. Note that max pain is a heuristic, not a target.

I have a call: [CALL]. Log it as a prediction with your TradeLoop wiki tools — extract the claim, my probability (%), the ticker, and a resolve-by date and resolution criterion; if I didn't give a probability or date, ask for them in one line first. Confirm what you logged. (Grade due ones later with resolve_prediction; see my calibration anytime.)

Run my pre-market prep with your TradeLoop tools:
1) SPY and QQQ: overnight/pre-market change, and any macro releases scheduled today.
2) For each of my watchlist names ([SYMBOLS]): gap % vs yesterday's close, prior-day high/low, and the daily confluence score.
3) Scan today's large-cap movers for anything I should know about outside my list.
Output: one table for the watchlist, then exactly 3 bullets titled "what matters today". Label data timestamps; research only, no trade calls.

Write me today's market brief using your TradeLoop tools: index moves (SPY/QQQ/IWM), the top sector story, the 3 biggest single-name stories on real volume (with the WHY, from news), and one macro data point that mattered. Every claim needs a source or a data timestamp. Separate clearly: FACT (prints, filings) vs INTERPRETATION (the market's read). No advice.

Run my end-of-day review with your TradeLoop tools:
1) How did my watchlist ([SYMBOLS]) close — % change, volume vs average, anything that broke a key level?
2) Did any of my monitoring alerts fire today?
3) One paragraph: what today's action means for my open theses.
Then write the review to my journal (wiki) as today's entry. Record only — no advice.

Run my weekend prep with your TradeLoop tools:
1) This week: how did SPY/QQQ and my watchlist ([SYMBOLS]) actually do — weekly % change and where each sits vs its 50-day MA.
2) Next week: which of my names report earnings, and what macro releases (CPI, Fed, jobs) are scheduled.
3) Regime check: is the tape trending or ranging (higher-timeframe read on SPY)?
Output a one-page plan: 3 things to watch, 2 risks, 1 discipline reminder. Write it to my journal. Research only.

Prep me for earnings this week using your TradeLoop tools: for each of my watchlist names ([SYMBOLS]) reporting within 7 days — report date/time, consensus EPS and revenue, the last 4 quarters' beat/miss record, and the options-implied expected move if available. Flag any name that's already stretched (overbought/oversold) into the print. Table + one caveat per reporting name. Research only — the hold/trim call is mine.

Read the crowd on [SYMBOL] using your TradeLoop tools: search X/FinTwit and Reddit for the past week's chatter, plus news headlines. Tell me: is the crowd leaning bullish or bearish, what's the dominant narrative, and — importantly — is the sample thin or skewed by one loud account? Cite the actual posts. Then contrast with what the price/volume data says. Sentiment = attention, not truth; no advice.

Scan for unusual volume with your TradeLoop tools: today's relative-volume leaders (volume vs their own recent average, min price $5, min dollar volume $20M). For the top 5, pull the news or filing that explains the attention. Table: symbol, RVOL, % change, one-line WHY (or 'no visible catalyst' — that's information too). Research only.

Screen my watchlist ([SYMBOLS]) for relative strength using your TradeLoop tools: each name's 21-day and 63-day excess return vs SPY, whether its RS line is rising, and its trend regime. Rank leaders to laggards. Output the ranked table plus one line: where is leadership concentrating? Research only.

Map sector rotation with your TradeLoop tools: rank the 11 sector ETFs (XLK, XLF, XLV, XLE, XLY, XLP, XLI, XLB, XLRE, XLU, XLC) by relative strength vs SPY over ~1 and ~3 months. Output the ranked table with both windows, then 2 lines: which rotation story the ranking tells, and whether it's risk-on or risk-off. Research only.

Run a macro regime check with your TradeLoop tools: latest 10y-2y curve spread and its 3-month direction, VIX level vs its recent range, the dollar index trend, the current Fed funds rate, and SPY's trend regime (trending/ranging, above/below the 200-day). Synthesize into ONE regime label (risk-on / neutral / risk-off) with the two strongest supporting facts and the one contradicting signal. Facts from data, label clearly marked as interpretation. No advice.

Run the daily crypto positioning read with your TradeLoop tools for [COIN]: funding rate (current + recent trend), open interest (is the money new or unwinding?), liquidation skew over the past day, an order-flow snapshot (book + tape lean), and spot price context (24h/7d change). One line per signal, then a combined read: is positioning crowded, and which direction would hurt the crowd most? Log the read to my journal. Research signal only — never a trade call.

Review my portfolio with your TradeLoop tools: current P&L per position and total, then for each holding answer four questions — (1) is the technical picture still consistent with why I own it, (2) how far is price from my cost and from the nearest key level, (3) any earnings within 14 days, (4) is any single position over 25% of portfolio value? End with the one position that most needs a decision this week and why. Research only — the decision is mine.

Do a visual chart review of [SYMBOL] with your TradeLoop tools: render the daily candlestick chart (SMAs, Bollinger, RSI, MACD), then read the STRUCTURE from the picture — trend shape, where price sits vs the moving-average stack, any base/breakout/breakdown pattern, and whether the bands are squeezing. Cross-check the visual read against the computed indicators and flag any disagreement. 5 bullets max. Research only.

Scan my watchlist ([SYMBOLS]) for volatility squeezes using your TradeLoop tools: for each name check whether the TTM squeeze is on, the Bollinger bandwidth percentile (tighter = more coiled), and the trend direction it would likely resolve toward. Rank by tightness. Output the table plus one line per squeezed name: what would confirm the fire (direction + volume). Research only, no entries recommended.

Do a full research pass on [SYMBOL] with your TradeLoop tools:
1) Price & technicals: current quote, trend regime, confluence score, key levels.
2) Fundamentals: next earnings + estimates, valuation snapshot, recent EPS trend.
3) Filings: the most recent 8-K/10-Q and any insider transactions.
4) Narrative: FinTwit/Reddit sentiment (cited) and this week's news.
5) Synthesis: the bull case, the bear case, and what would make each wrong.
Every claim labeled: fact (filing/print) vs interpretation vs speculation. Save the writeup to my journal. Research only — no recommendation.

Use your TradeLoop scanner (fundamental_screen) to find US large caps with market cap ≥ $10B, P/E ≤ 20 and dividend yield ≥ 2.5%; then read ROE and debt/equity on the top rows. Return a ranked table and say which rows are fully enriched. Research pool, not picks.

Use your TradeLoop scanner (near_52w_low) for S&P 500 names within 10% of their 52-week low with market cap ≥ $10B. For the top 5, add a one-line fundamentals sanity check (P/E, yield, why it might be down). A watch pool, not a buy list.

Use your TradeLoop tools (edgar pe_band) to build [SYMBOL]'s 5-year trailing-P/E band: current P/E, min/median/max and today's percentile. One paragraph on what that regime implies — label facts vs interpretation.

Use your TradeLoop portfolio tools: pnl with benchmark SPY over the last 3 months (portfolio vs benchmark vs excess return), then risk() for correlations and beta. Name the biggest contributor and the biggest drag; note the current-weights assumption.

Run [SYMBOL] through a Buffett-style quality checklist with your TradeLoop tools (playbook: buffett-quality-check): pull the long-run financials via edgar company_concept and yahoo-finance fundamentals, compute owner earnings (net income + D&A − capex) and its trend, check ROE/margin durability and debt, check insider behavior via edgar insider_trades, then compare the owner-earnings yield to the 10y treasury (fred DGS10). Verdict format: 'quality: pass/fail on which tests' and 'price: what discount (if any) exists' — separately. No recommendation.

Run [SYMBOL] through a Lynch-style GARP framework with your TradeLoop tools (playbook: lynch-garp-screen): classify it (slow grower / stalwart / fast grower / cyclical / turnaround / asset play), compute the PEG from yahoo-finance fundamentals + the finnhub earnings trend, check the balance sheet, and tell the two-minute story — what it does, why earnings grow, what breaks it. Flag any classic GARP red flags (diworsification, hot-industry, whisper-stock). Research only.

Place the US economy on the growth × inflation regime map (Dalio-style) with your TradeLoop fred tools (playbook: dalio-regime-map): growth direction (GDPC1, UNRATE, PAYEMS momentum), inflation direction (CPIAUCSL, T5YIE vs target), the policy overlay (fed funds path, yield_curve inversion, net_liquidity trend). Name the quadrant, what historically works there, and — if I share my portfolio — which quadrant my current book is implicitly betting on. Environment classification, not market timing.

Take the market's temperature (Marks-style cycle read) with your TradeLoop tools (playbook: marks-market-temperature): sentiment dashboard (F&G + components), VIX percentile and term structure, risk-appetite tells (junk-bond demand component, what perplexity finds on IPO/speculative activity), and positioning (put/call, notable short interest). Score each gauge hot/neutral/cold, give an overall temperature, and state the posture it implies (more aggressive / neutral / more defensive) — posture only, no trades. Remind me extremes are the only actionable readings.

Use your TradeLoop tools to compare [SYMBOL] against its 3-4 closest peers (same industry). Pull fundamentals for each and build ONE table: market cap, forward P/E, PEG, revenue growth, gross & operating margin, ROE, net cash/debt. End with two lines — where [SYMBOL] looks cheap/expensive vs the group, and which peer screens best on quality-at-a-price. Facts vs interpretation labeled. Research only.

Use wiki_search / wiki_read to find my saved notes on [SYMBOL] (the ticker page and recent research). Read them in full, then re-verify each key claim against LIVE data — current price vs the levels I noted, latest earnings/estimates, any new filing, and whether the catalyst still stands. Report: what still holds, what's now stale or wrong, and whether my invalidation level has been hit. If I have no notes yet, say so. Research only — the decision is mine.

Use your TradeLoop portfolio + fundamentals tools: for each dividend-paying holding, pull the yield, annual per-share dividend and next ex-dividend / pay date, then compute my estimated annual income per position and portfolio total, and my portfolio-level yield on current value. List upcoming ex-dates in date order so I don't miss one. Flag any holding whose dividend looks stretched (very high yield / thin coverage) as to-verify. Research only.

Use your TradeLoop history tools on [SYMBOL]: compute average return and hit rate for the current calendar month over the last ~10 years (seasonality), the 10-year CAGR and the worst annual drawdown so I know the downside, and note whether the stock is currently at an unusual extreme vs its own history (distance from 200-DMA, RSI). Frame as context/base rates, not a forecast — label clearly. Research only.

It's after the open — run the opening-range check on [SYMBOL] with your TradeLoop tools:
1) Classify the overnight gap (size %, into resistance or out of a base?).
2) The 15-minute opening range: high/low, and where price is now relative to it.
3) Participation: RVOL now, and which side of the open's VWAP price is holding.
Then tell me which of the three cases this is — BREAK (range cleared with RVOL ≥ 1.5 and VWAP agreement), FADE (emotional gap stalling at a level on fading volume), or STAND DOWN (chop: repeated VWAP crosses, RVOL < 1) — with the evidence for each criterion. Use during regular hours. Research only: the entry, the fade, and the pass are all my call.

Run a VWAP read on [SYMBOL] (intraday, use during regular hours) with your TradeLoop tools: current price vs today's VWAP, how many times price has crossed VWAP this session (chop warning), RVOL, and whether money flow (CMF) agrees with the side price is on. Verdict: holding above (buyers control) / rejected at (sellers control) / churning through (no signal) — with the numbers behind it. Research only.

[SYMBOL] looks like it's breaking out — vet it with your TradeLoop tools before I act:
1) Is it actually clearing a real level (new Donchian high / above prior resistance)?
2) Participation: RVOL ≥ 1.5? OBV and CMF confirming?
3) Higher timeframe: does the weekly agree with the daily?
4) Fuel & risk: short interest % of float, and is it already overbought/exhausted?
Score each check pass/fail with the number, then an honest overall read: confirmed / unconfirmed / trap risk. Research only — no entry recommendation.

Check whether [SYMBOL]'s recent breakout is failing, using your TradeLoop tools: did price clear a real level and then close back inside it? What did volume do on the break vs the failure? Is OBV diverging from price? Where would the trap thesis be wrong (the level to watch)? Conclusion: intact / failing / confirmed failure, with the specific closes and levels as evidence. Research only.

[SYMBOL] looks washed out — check for a REAL reversal setup with your TradeLoop tools: RSI level AND whether it's diverging bullishly from price, position vs the lower Bollinger band, whether the regime is a range (mean-reversion works) or a downtrend (falling knife), and whether any reversal candle has volume behind it. Honest verdict: confirmed reversal setup / just oversold (no confirmation) / downtrend — do not fade. List what confirmation is still missing. Research only.

Evaluate a mean-reversion fade on [SYMBOL] with your TradeLoop tools. Gate first: is the regime RANGING (low ADX, choppy)? If it's trending, stop and say so — fading trends is how accounts die. If ranging: how stretched is price (Bollinger %B, Williams %R, RSI), is there a divergence supporting the turn, and where are the range boundaries the trade would target and invalidate at? Verdict with numbers. Research only — sizing and trigger are mine.

Check whether [SYMBOL] is at a valid trend-pullback spot using your TradeLoop tools: is the trend actually established (MA stack aligned, ADX > 20, higher timeframe agreeing)? Has price pulled back to a logical level (20/50-day MA or prior breakout zone) without breaking trend structure? Is RSI resetting to the 40-50 zone rather than collapsing? Is selling volume drying up on the dip? Verdict: valid pullback / trend breaking / not there yet — each criterion with its number. Research only.

Help me size a [SYMBOL] position by risk, using your TradeLoop tools for the volatility inputs. My account is [ACCOUNT SIZE] and I risk [RISK %] per trade. Pull the current ATR(14) and recent volatility, propose a stop distance (e.g. 1.5-2× ATR below entry for a long — show both), and compute: risk dollars, position size in shares for each stop choice, and what % of the account that position is. Flag if the name's volatility makes the position uncomfortably large or small. Math only — where to actually enter and whether to trade at all is my call.

I'm long [SYMBOL] from [ENTRY PRICE] and it's working. Using your TradeLoop tools, build me the management picture: current price and open gain %, a chandelier/ATR trailing stop level (show the math), whether the trend is still healthy (MA stack, higher-timeframe agreement) or showing exhaustion (overbought + divergence), and the level where the trend thesis would actually break. Lay out the standard options (trail full / scale a third at exhaustion signs / exit on structure break) with what each costs if the trend continues. Analysis only — the decision is mine.

Help me frame the earnings decision on [SYMBOL] (report within days) using your TradeLoop tools: the options-implied expected move for the report, how the stock ACTUALLY moved after each of its last 4 reports (gap % day after), its beat/miss record, and current historical volatility. Then frame it: if I hold through, what's the realistic downside case at my position size — and what does the base rate say about post-print direction for this name? Numbers and base rates only — hold / trim / exit is my decision, and say exactly that.

Run the full order-flow read on [COIN] with your TradeLoop tools: sample the Order Flow Imbalance over ~20 seconds, take a book-imbalance snapshot (top 10 levels), and read the trade-sign imbalance from the recent tape. Apply the agreement rule: only call pressure directional when book AND tape lean the same way; otherwise call it mixed. Add funding + OI for the positioning backdrop. This is a SHORT-HORIZON research signal (minutes) — say so in the output, and no trade call.

Run an adversarial deep dive on [SYMBOL] with your TradeLoop tools (playbook: bull-bear-deep-dive):
1) Gather in parallel: fundamentals (edgar/yahoo-finance), technicals (indicators.analyze, day, benchmark SPY, htf week), crowd positioning (x-search + short interest), insider activity (edgar), and the macro backdrop (fred yield_curve + net_liquidity, sentiment dashboard).
2) Argue the strongest honest BULL case — every claim cites a number.
3) Argue the strongest honest BEAR case — no strawmen, find the real one.
4) Verdict as a CONDITIONAL, not a rating: 'bull case holds unless X, bear case activates if Y'.
5) Save the debate to my wiki journal.
Label fact vs interpretation vs speculation throughout. Research only — no buy/sell call.

I'm long [SYMBOL] from [ENTRY PRICE] and it's underwater. Using your TradeLoop tools, build the cut-or-hold picture WITHOUT anchoring on my loss: 1) read my saved thesis/notes (wiki) — what did I actually expect and what was my invalidation level? 2) has that level been hit, or is price still inside the noise? 3) is the technical damage structural (broke a real level, MA stack rolled over, HTF turned) or cosmetic? 4) any thesis-breaking news/filing since I bought? Lay out cut / hold / add-here with the condition that decides each. Be honest about sunk cost. Research only — the decision is mine.

Analyze [SYMBOL] with your TradeLoop tools using the disruptive-innovation lens (playbook: disruptive-innovation-lens). Run the three thesis tests (Wright's-Law cost curve, S-curve penetration, TAM expansion), then the condensed memo: value-chain seat and whether it drives vs leverages the platform, moat, leadership (founder-led / technical CEO?), financials and cash runway, valuation and PEG, consensus upside, thematic-fund alignment (label to-verify), and the fragility flags. Then compute the 0-125 Technology-Valuation Score with the A/B/C breakdown and give the rating (STRONG BUY 100-125 / BUY 80-99 / HOLD 60-79 / SELL <60), a conviction, a valuation view, and 1-year bull / base / bear targets. Show both bull and bear before the rating, label each claim fact vs interpretation vs speculation, end with the Innovation-Fit Summary. No buy/sell advice — research only.

Suggest 3-5 disruptive-innovation names in [THEME] that fit the disruptive-innovation philosophy (S-curves, Wright's-Law cost deflation, platform convergence) using your TradeLoop tools (playbook: disruptive-innovation-lens). Screen out anything that fails the three thesis tests or trips the fragility flags (needs cheap capital, dilution-funded, narrative-only). For each survivor: a one-line thesis, where it sits on the S-curve, its convergence points, and a quick Technology-Valuation Score with rating. Rank by score, and give a bull and a bear case for the top pick. These are ideas for research, not buy calls — label facts vs interpretation.

Run a Serenity-style bottleneck scan on [THEME] with your TradeLoop tools. First call serenity framework for the method, then: translate the theme into a system change and its binding physical constraint; map the value chain and RANK THE LAYERS by scarcity before naming companies; build a 20+ candidate universe across layers (perplexity for the narrative sweep, edgar filing_search for who names the technology in real filings, cn-market for A股 legs); grade evidence by the ladder (filings > trade press > KOL threads = leads only); score the 3-7 finalists with serenity scorecard; state each one's exact chokepoint, the strongest evidence, and the kill switch. Include one popular name that ranked LOWER and why. 开头先给层排序（'Start with the layers…'），结论是优先研究名单不是买入清单。

Challenge [SYMBOL] as a supply-chain chokepoint with your TradeLoop tools (serenity framework for the method). Determine its EXACT value-chain position and classify honestly: controls the scarce layer / supplies it / benefits from the trend / mainly a story. Pull hard evidence — edgar filing_search for customer concentration and capacity language, insider_trades, batch_fundamentals vs the closest peers, perplexity for the current narrative — grade each source by the evidence ladder, then run serenity scorecard and report the per-factor breakdown. End with: what the market may be missing, the 2-3 kill switches, and the single next check that would most change the answer. 判断要敢下，但每个论断标注证据强度。研究支持，不是买卖建议。

Value [SYMBOL] with your TradeLoop tools — a scenario-based intrinsic estimate, not a price target to trade.
1) Pick the method by lifecycle and say why: loss-making/early → P/S primary; mature & profitable → P/E + PEG; asset-heavy/financial → P/B + ROE.
2) Pull inputs — revenue, net margin, EPS trend, share count, free cash flow, growth and consensus estimates (edgar company_concept, yahoo-finance fundamentals, finnhub estimates); anchor to the fiscal year covering the next 12 months.
3) Build THREE independent scenarios — bear / base / bull — each with its OWN revenue-growth and net-margin assumption → its own forward EPS × a scenario-appropriate multiple = a fair-value estimate. Never one EPS times three multiples.
4) If free cash flow is positive, add a simplified DCF: grow FCF 5 years at your stated rate, a perpetuity terminal value, discounted at 10% (cite the 10y via fred DGS10 as the risk-free anchor). Reconcile it against the multiple-based range.
5) Output a bear/base/bull fair-value RANGE with the EPS and multiple printed for each, the implied up/downside vs the current price, the margin of safety at today's price, and the single assumption the range is most sensitive to.
Label every input fact (filing) vs interpretation vs speculation; mark consensus numbers 'to-verify'. Research only — a valuation range under stated assumptions is not advice and not a call to buy or sell.

Produce a full institutional-style research report on [SYMBOL] with your TradeLoop tools. Gather in parallel, then write these sections:
1) Core read (3-4 lines): what it is, the one thing that matters most, current price and trend regime.
2) Business & moat: what it sells, competitive position, and whether the advantage is durable.
3) Financials & trend: revenue / margin / EPS / free-cash-flow trajectory and balance-sheet health (edgar company_concept, yahoo-finance).
4) Technicals: trend, confluence score, key levels (indicators.analyze, levels) — render a chart.
5) Sentiment & consensus: FinTwit/Reddit (x-search, cited), analyst consensus (finnhub recommendation), and this week's news.
6) Peers: one row per peer on valuation / growth / margins.
7) Risk & poke-holes — the strongest HONEST bear case: for each key business line screen for tech obsolescence, customer concentration, competitive pressure, cycle-peak earnings, supply-chain fragility, and pricing power, and test whether profit data (not just the story) supports the moat; grade each defect fatal / structural / severe / moderate / mild.
8) Valuation: bear/base/bull fair value from INDEPENDENT scenario EPS × multiple, plus a simple DCF if FCF is positive, every assumption printed (fair value, not a price target).
9) Research score — score each 0-10 and show the numbers: long-term = industry position, growth certainty, valuation reasonableness, moat; short-term = technical setup, near-term catalysts, flow (volume/RVOL trend — say so if institutional-flow data isn't available, don't guess), and reward-vs-risk. Report every sub-score plus the two arithmetic-mean composites.
10) Risk-reward map: place it on the odds (confidence in the base case) × payoff (discount to fair value) grid and name the cell descriptively (asymmetric-and-confident / speculative / priced-in / unattractive) — a characterization, not a position instruction.
11) Conditional verdict + watch list: 'the base case holds unless X; the bear activates if Y', then 5 metrics to monitor and how often to review each.
Label every claim fact (filing/print) vs interpretation vs speculation; treat consensus and social as to-verify. Save the report to my wiki journal. Research only — no recommendation, no position sizing, no buy or sell call.

Set up price alerts with your TradeLoop monitor tools: watch [SYMBOL] and alert me if price breaks above [UPPER LEVEL] or below [LOWER LEVEL]. Register the rules, then read back exactly what's now being watched so I can confirm. These persist and are checked whenever a monitoring sweep runs — also tell me how to trigger a sweep or put one on a schedule.

Arm indicator alarms on my watchlist ([SYMBOLS]) with your TradeLoop monitor tools: for each name — RSI(14) below 30 or above 70, a MACD signal cross in either direction, and relative volume above 2×. Use replace mode so stale rules don't accumulate. Read back the full rule set per symbol when done. Alerts are edge-triggered (they fire once per crossing, not continuously) — confirm that's the behavior I want or adjust.

Set up my recurring market watch: every 5 minutes during market hours, run a TradeLoop monitor sweep. When anything fires: 1) push a desktop banner with the alert text, 2) append one line to my journal (symbol, trigger, price, time), 3) if it's a price-level break, add one sentence of context (RVOL + nearest key level). When nothing fires, stay silent. Use a recurring loop (e.g. /loop 5m) — set it up now and tell me how to stop it later.

Cadence guidance: 5 min ≈ 78 sweeps/session. History is served from local cache; each sweep costs ~1-2 market-data requests thanks to batched snapshots.

Guard my holdings with your TradeLoop monitor tools: for every symbol in my portfolio, register two rules — alert me when earnings are within 3 days, and alert me on any NEW SEC filing (the filing rule arms silently on the current latest and only fires on genuinely new ones). Read back what's armed. Then tell me how these get checked — I want them included in my recurring monitoring sweep.

Automate my morning briefing: schedule a recurring task for 9:00 AM ET on weekdays that runs my pre-market prep (watchlist gaps & key levels, macro calendar, large-cap movers), writes the brief to my journal, and pushes a desktop banner with the single most important line. Set the schedule up now (use a scheduled/recurring agent run), show me exactly what will execute, and tell me how to pause it.

One run per weekday ≈ 10-15 market-data requests — negligible against monthly quota.

Put my trading journal on autopilot: schedule a recurring task for 4:15 PM ET on weekdays that (1) pulls how my watchlist and portfolio closed, (2) checks which monitor alerts fired today, (3) writes a dated end-of-day entry to my journal (wiki) with those facts plus a one-paragraph synthesis, and (4) if any position moved more than 3%, flags it for my review tomorrow. Record only — the entry states facts and open questions, never advice. Set it up and show me the first entry format.

Set up my portfolio risk watch with your TradeLoop tools: 1) register a monitor rule that alerts when ANY position grows past [MAX WEIGHT]% of portfolio value (use the portfolio-weight rule on '*'); 2) schedule a weekly Friday-afternoon risk report — per-position weights, portfolio volatility and drawdown for the recent period, and the two most correlated positions — written to my journal. Arm both now and read back what's active. The point is deciding tripwires while calm, not in a drawdown.

Use TradeLoop monitor: add a portfolio_weight_above rule on symbol '*' with value [LIMIT] so a check fires once whenever any position crosses that weight. Then run a check now and tell me the current heaviest weights.

Pairs with the desktop's background monitor loop — it re-checks every few minutes.

I'm about to act on [SYMBOL]. Before anything else, write a decision-journal entry to my wiki (playbook: decision-journal-calibration): 1) my thesis in one sentence + edge category, 2) entry/target/stop/size and R:R as I state them, 3) my stated confidence % that this wins, 4) the kill switch — what invalidates it before the stop, 5) one-word state check. Snapshot the current indicators.analyze read (confluence, regime) into the entry. Then ask me the two fields I left vague. Do not evaluate the trade — just record it faithfully.

Run my calibration review with your TradeLoop wiki tools. FIRST call calibration for my prediction ledger — Brier score, hit rate vs mean confidence, and where each probability bucket lands (are my 70% calls hitting ~70%?). THEN search my journal for the month's decision entries and add: expectancy by edge category (what actually pays me), a bias tally (kill-switch violations, moved stops, FOMO entries), and any predictions still due to grade. End with the ONE behavioral goal for next month, from my own data. Be brutally honest; don't rationalize wrong calls as 'early'. Write the review back to the wiki.

Automate a disruptive-innovation watch: schedule a recurring weekday task that re-scores my innovation holdings [SYMBOLS] with the disruptive-innovation lens (playbook: disruptive-innovation-lens). Each run: refresh price, valuation/PEG, consensus target, and any news catalyst, recompute the 0-125 Technology-Valuation Score, and flag any name whose rating changed or whose thesis milestones (cost/unit, penetration, gross-margin path recorded in my wiki) broke. Write the summary to my journal and push a desktop banner ONLY when a rating flips. Set the schedule up now, show me exactly what will run, and tell me how to pause it.

One re-score pass per weekday across a handful of names — light against monthly quota.