opencode-skills-collection 4.0.73 → 4.0.75

package/bundled-skills/traderspy-technical-analysis/SKILL.md

@@ -0,0 +1,162 @@
+---
+name: traderspy-technical-analysis
+description: "Read one crypto futures pair with TraderSpy: 19 indicators on up to 3 timeframes in one call, key levels, funding, open interest and positioning. Use for \"analyse BTC\" or \"is SOL oversold\"."
+category: finance
+risk: safe
+source: "https://github.com/target1m/traderspy-mcp/tree/069aae5a84640d671f92705a2e3bc0b62efca1e0/skills/technical-analysis"
+source_repo: target1m/traderspy-mcp
+source_type: official
+date_added: "2026-09-25"
+author: target1m
+tags: [traderspy, crypto, technical-analysis, indicators, mcp]
+tools: [claude, cursor, gemini]
+license: MIT
+license_source: "https://github.com/target1m/traderspy-mcp/blob/069aae5a84640d671f92705a2e3bc0b62efca1e0/LICENSE"
+---
+
+# TraderSpy Technical Analysis
+
+One good read of a chart is three things in order: where price is in its structure (trend and
+levels), what momentum and volatility are doing, and how the derivatives market is positioned
+around it. TraderSpy gives all three in at most two calls. Spend the calls well — the user's quota
+is finite — and quote the numbers, not adjectives.
+
+## When to Use
+
+- Use when the user asks about a coin's chart, trend, momentum, support and resistance, or where a level such as the 200 EMA sits.
+- Use for "analyse BTC", "how does SOL look", "is X overbought / oversold", "what do the indicators say" or "is this a good entry", even when the user only names a coin.
+- Use when the question is whether funding is high or open interest is building, or when the user wants price or candle data.
+- For scanning many coins by conditions use `@traderspy-market-screener`; for whale positions use `@traderspy-smart-money`.
+
+## Tools
+
+| Tool | Use it for | Key arguments |
+| --- | --- | --- |
+| `get_technical_indicators` | The chart read | `symbol`, `intervals` (≤ 3) or `interval`, `indicators` (≤ 19), `periods`, `history` 0–20 |
+| `get_derivatives` | Funding, open interest, positioning | `symbols` (≤ 5 Binance USDⓈ-M perps; bare `BTC` is fine) |
+| `get_price` | Live price, 24h high/low/volume/change | `symbols` (≤ 20) |
+| `get_candles` | Raw OHLCV | `symbol`, `interval` 1m/5m/15m/1h/4h/1d, `limit` ≤ 500 |
+| `get_tracked_symbols` | Coverage check | none — only when a symbol returns no data |
+
+Symbols are Binance Futures names: `BTCUSDT`, `SOLUSDT`, `1000PEPEUSDT` (1000x-prefixed small
+caps), and tokenized equities/ETFs such as `AAPLUSDT` or `SPYUSDT`. Append `USDT` to a bare coin.
+Hyperliquid-only perps (e.g. HYPE) are tracked under the same naming.
+
+## Making the call count
+
+- **Multi-timeframe is one quota unit.** For any "how does X look" question pass
+  `intervals: ["1h","4h","1d"]` (or `["15m","1h","4h"]` for an intraday user) and get a
+  `confluence` verdict — never three separate calls.
+- **Pick indicators for the question.** Default `rsi, macd, ema, bollinger` covers momentum and
+  trend. Add `levels` + `pivots` whenever the user cares about entries, targets or "where is
+  support"; `atr` for stop distance and volatility; `adx` for trend strength; `supertrend` for a
+  clean trend state; `obv` / `mfi` when volume confirmation matters; `ichimoku` / `keltner` only
+  when asked. Eight to ten well-chosen indicators is a full read; all 19 is noise.
+- **`periods`** lets you honour the user's settings (`{"ema":[9,21,55,200],"rsi":7}`); an unknown
+  key is rejected, so use the documented names only.
+- **Add `get_derivatives`** when the question involves "why is it moving", crowding, funding,
+  open interest, leverage, liquidations, or when the user is about to lean on a directional call —
+  candles cannot see positioning.
+- `get_candles` is for when the user wants the data itself (to chart, export or compute something
+  custom). Do not fetch candles to eyeball indicators the technicals tool already computes.
+
+## Reading the response
+
+**Shape.** One timeframe → top-level `indicators` + `summary`. Several → `timeframes[]` (each
+with `interval`, `price`, `candles`, `indicators`, `summary`) plus `confluence` `{ aligned, bias,
+byInterval }`; the top-level `interval`/`price` then name the first timeframe only. `settings`
+echoes what was computed; `warnings[]` lists anything that was thinned or missing.
+
+**Every indicator** carries `value`, `previous`, `direction` (rising / falling / flat) and a short
+`series` (oldest → newest), plus its own fields — `zone` on RSI/Stochastic/CCI/MFI/Williams,
+`crossover` on MACD/Stochastic, `stack`/`pricePosition` on EMA, `±DI` and `strength` on ADX,
+`trend`/`flipped`/`barsSinceFlip` on SuperTrend, `divergence` on RSI/MACD/OBV. `{ value: null,
+reason: "insufficient_data" }` means the tape is too short for that period — say so, do not
+substitute.
+
+**`summary` is the spine of the read.** `bias` (bullish / bearish / neutral) is computed from a
+FIXED set (RSI 14, MACD, EMA 20/50/200, ADX 14, ATR 14, BB 20/2, volume 20) regardless of what
+you requested, so it is comparable between calls; `score` is its internal vote (|score| ≥ 30 sets
+a bias). `trend` (direction, strength, emaStack, priceVsEma200), `momentum` (rsiZone, macdState),
+`volatility` (atrPct, atrPercentile 0–100, squeeze, state), `volume.ratioVsAverage`, `patterns[]`
+(candlestick finders on the last two bars) and `notes[]` — plain-language sentences written to be
+quoted. Build the answer from the notes, then add the numbers.
+
+**`levels`** (nearest three per side from swing highs/lows over 300 bars): `price`, `touches`,
+`distancePct`; `range.positionPct` (0 = at the low of the lookback, 100 = at the high);
+`fibonacci` (the dominant leg with `retracedPct` and the nearest level above/below);
+`volumeProfile` (`poc`, value area, `pricePosition`). **`pivots`** are classic floor pivots from
+the previous DAILY candle whatever the timeframe. Together they give you a levels table; distance
+in % is what a user can act on.
+
+**`confluence.aligned`** is true only when every timeframe with data agrees on a non-neutral bias.
+"Aligned bearish on 1h/4h/1d" is a strong statement; "4h bullish, 1d neutral" is a pullback
+question. Say which.
+
+**Derivatives** (`get_derivatives`): `funding.label` — `extreme_longs_paying` / `longs_paying` /
+`neutral` / `shorts_paying` / `extreme_shorts_paying` (longs pay at ≥ +0.03% per 8h, extreme at
+≥ +0.10%; shorts pay at ≤ −0.01%, extreme at ≤ −0.05%; Binance's baseline is +0.01%; quote
+`annualizedPct`); `openInterest.regime` — `new_longs` / `short_covering` / `new_shorts` /
+`long_liquidation` / `flat` (needs ≥ 1% OI and ≥ 0.5% price change over 24h);
+`positioning.label` — `long_heavy` (long/short ratio ≥ 1.5, i.e. ≥ 60% long) / `balanced` /
+`short_heavy` (≤ 0.67, i.e. ≤ 40% long), read from the top-trader POSITION ratio first;
+`takerBuySellRatio` is the last hour's aggressor flow. Data is at most 60 s old. `notes[]` are
+quotable. Details in `references/derivatives-guide.md`.
+
+## Presenting
+
+For a single-coin read ALWAYS use this order, and keep it to what the user asked:
+
+1. **Price line** — symbol, price, time of the read, 24h change if you fetched it.
+2. **Per timeframe, one line each** — bias, trend (EMA stack and ADX), momentum (RSI zone, MACD
+   state), volatility (ATR %, squeeze), with the key number after each label.
+3. **Confluence** — aligned or not, and what the disagreement means.
+4. **Levels table** — nearest support and resistance with distance %, the pivot, the nearest
+   Fibonacci level, the value-area edge if price is near it.
+5. **Derivatives line** (if fetched) — funding label + annualized, OI regime, top-trader long %.
+6. **What would change the read** — the one or two levels or readings that flip the picture
+   (e.g. "a 4h close above 79,325 resistance with OI rising would turn the 4h bias bullish").
+7. One plain sentence: crypto derivatives are high-risk; this is market information, not advice.
+
+Use `RSI(14) 41.3 — neutral, falling` style labelling so the period and the direction are always
+visible. Percentages to one decimal; prices to the precision the tool returned.
+
+## Conduct
+
+- Describe the chart; do not prescribe the trade. If the user asks "is this a good entry", answer
+  with what supports the entry, what argues against it, where the invalidation sits, and hand the
+  decision back. Never tell the user to buy, sell, size or leverage.
+- Indicators describe the past bars they were computed on; do not present a bias as a forecast or
+  imply any outcome is assured.
+- Nothing in this connector trades — no order, close, transfer or withdrawal tool exists, by
+  design. If asked to execute, say so plainly; the decision and the trade stay with the user.
+- Quote only what the tools returned. If an indicator is `insufficient_data` or a symbol is not
+  tracked, say so instead of estimating.
+- When the answer is about a specific trade idea, end with one plain sentence that crypto
+  derivatives are high-risk and this is market information, not financial advice — once per
+  answer.
+
+Per-indicator fields and conventional readings: `references/indicator-guide.md`.
+
+## Examples
+
+```text
+Analyse BTC on the 1h, 4h and 1d with funding and open interest.
+Is SOL overbought? Where are the nearest support and resistance levels?
+Where is the 200 EMA on ETH, and is funding stretched?
+```
+
+## Limitations
+
+- Needs the hosted TraderSpy MCP server connected in the client (`https://mcp.traderspy.app/mcp`, Streamable HTTP), authorized with OAuth or a personal key from https://traderspy.app/mcp. A free TraderSpy account is enough. Without the server the skill has no data to work from.
+- Tool calls are metered per day: 300 on the free tier, 5,000 on premium.
+- Every tool is read-only. Nothing here places, closes or modifies an order, and there is no withdrawal or transfer tool.
+- The output is market data and analysis for the user's own research, not investment advice.
+- A multi-timeframe read is one tool call; adding derivatives makes it two.
+- Symbols use Binance Futures naming (`BTCUSDT`, `1000PEPEUSDT`). `get_derivatives` covers Binance USDⓈ-M perpetuals, up to five per call.
+
+## Related Skills
+
+- `@traderspy-market-screener` - many coins at once, and backtests of a condition
+- `@traderspy-smart-money` - what tracked top traders hold
+- `@traderspy-position-check` - the same read applied to an open position

package/bundled-skills/traderspy-technical-analysis/references/derivatives-guide.md

@@ -0,0 +1,57 @@
+# Derivatives guide — `get_derivatives`
+
+Source: Binance USDⓈ-M perpetual futures public endpoints, snapshot cached 60 s. One entry per
+symbol; an unknown or non-Binance symbol comes back as `{ symbol, error, message }`.
+
+## Fields
+
+- `markPrice`, `indexPrice`, `premiumPct` — mark vs index; a persistent positive premium is
+  another face of long crowding.
+- `funding`: `rate` (raw, per 8h), `ratePct`, `annualizedPct`, `nextFundingTime`,
+  `minutesToNextFunding`, `avg24hPct`, `avg3dPct`, `label`.
+- `openInterest`: `contracts`, `valueUsd`, `change4hPct`, `change24hPct`, `priceChange24hPct`,
+  `regime`.
+- `positioning`: `globalLongShortRatio`, `globalLongPct` (all accounts), `topTraderLongShortRatio`,
+  `topTraderLongPct` (top traders, by position size), `takerBuySellRatio` (last hour),
+  `label`.
+- `notes[]` — the interpretation in sentences; quote them.
+
+## Labels
+
+| `funding.label` | Rule (per 8h) | Reading |
+| --- | --- | --- |
+| `extreme_shorts_paying` | ≤ −0.05% | Shorts crowded and paying heavily; squeeze risk |
+| `shorts_paying` | ≤ −0.01% | Shorts paying longs |
+| `neutral` | between −0.01% and +0.03% (Binance's baseline is +0.01%) | No crowding signal |
+| `longs_paying` | ≥ +0.03% | Longs paying shorts — leverage leaning long |
+| `extreme_longs_paying` | ≥ +0.10% | Longs crowded and paying heavily; long-squeeze / flush risk |
+
+Quote `annualizedPct` alongside — "0.05%/8h" means little to most readers, "≈ 55% APR" does.
+
+| `openInterest.regime` | OI 24h | Price 24h | Reading |
+| --- | --- | --- | --- |
+| `new_longs` | up | up | Fresh money buying; trend being funded |
+| `short_covering` | down | up | Rally driven by shorts closing — less durable |
+| `new_shorts` | up | down | Fresh shorts pressing; trend being funded to the downside |
+| `long_liquidation` | down | down | Longs being flushed — capitulation-type move |
+| `flat` | < 1% OI or < 0.5% price | No meaningful OI story |
+
+| `positioning.label` | Rule | Reading |
+| --- | --- | --- |
+| `long_heavy` | long/short ratio ≥ 1.5 (≈ 60%+ long) | Leaders leaning long |
+| `balanced` | between | No lean |
+| `short_heavy` | ratio ≤ 0.67 (≈ 40% or less long) | Leaders leaning short |
+
+Top-trader positioning (by position size) is read first because it is the informed cohort; the
+all-account ratio shows the crowd. When they disagree ("top traders 69% long, all accounts 45%
+long"), that disagreement is the interesting sentence.
+
+## Combining with the chart
+
+- Chart bullish + `new_longs` + `neutral` funding: a trend with fuel and no crowding.
+- Chart bullish + `extreme_longs_paying`: extended; flush risk is the caveat to state.
+- Chart bearish + `short_covering` bounce: a relief rally, not a reversal, until OI rebuilds.
+- `long_liquidation` with `extreme_shorts_paying` afterwards: the flush may be done; watch whether
+  OI rebuilds.
+
+These are conventional readings to explain the data, not predictions.

package/bundled-skills/traderspy-technical-analysis/references/indicator-guide.md

@@ -0,0 +1,47 @@
+# Indicator guide — fields and conventional readings
+
+Every indicator object carries `value`, `previous`, `direction` (rising / falling / flat) and
+`series` (history + 1 values, oldest → newest). Extra fields per indicator below. "Reading" is
+the conventional interpretation, not a prediction.
+
+| Key | Extra fields | Reading |
+| --- | --- | --- |
+| `rsi` | `period`, `zone` (oversold < 30, neutral, overbought > 70), `divergence` `{type, barsAgo, priceFrom, priceTo, oscillatorFrom, oscillatorTo}` or null | 40–60 is neutral; a rising RSI from 30 is recovering momentum, not "still oversold". A bearish divergence (price higher high, RSI lower high) within the last ~60 bars is worth one sentence |
+| `macd` | `periods {fast, slow, signal}`, `macd`, `signal`, `histogram`, `previous {…}`, `crossover` (bullish / bearish / null — on the latest bar only), `divergence` | Histogram sign + direction is the state (`summary.momentum.macdState`: positive_rising, positive_falling, negative_rising, negative_falling). A cross on the latest bar is an event; report it as such |
+| `ema` / `sma` | `values[] {period, value, previous, direction, priceAbove, distancePct, series}`, `stack` (bullish = fast > mid > slow, bearish, mixed), `pricePosition` (above_all / between / below_all) | Stack + price position is the trend structure. `distancePct` from the 200 is the "how extended" number |
+| `bollinger` | `upper`, `middle`, `lower`, `bandwidthPct`, `percentB`, `position` (above_upper / upper_half / lower_half / below_lower), `squeeze` | %B > 1 or < 0 = outside the bands; `squeeze` true = bands inside the Keltner channel (TTM) — expect expansion, direction unknown |
+| `atr` | `period`, `atrPct` | Volatility in price units and as % of price. A stop tighter than ~1 ATR is inside noise on that timeframe |
+| `adx` | `plusDI`, `minusDI`, `strength` (absent / weak / strong / very_strong), `diBias` (bullish when +DI > −DI) | ADX measures strength, not direction; the DI bias gives the direction |
+| `stochastic` | `k`, `d`, `zone`, `crossover` | < 20 oversold, > 80 overbought; a %K/%D cross inside a zone is the classic trigger |
+| `obv` | `average`, `slopePct`, `divergence` (bullish / bearish / none) | Volume confirmation. `divergence: bullish` = price fell but OBV rose (accumulation). Use the direction and the divergence, not the raw level, which depends on where the window starts |
+| `vwap` | `period`, `priceAbove`, `distancePct` | Rolling VWAP over `period` bars (default 48); above = buyers in control over that window |
+| `cci` | `zone` | > 100 / < −100 = extended |
+| `mfi` | `zone` | Volume-weighted RSI; < 20 / > 80 |
+| `williamsR` | `zone` | −100…0; < −80 oversold, > −20 overbought |
+| `roc` | `period` | % change over the period; sign and direction |
+| `supertrend` | `trend` (up / down), `previousTrend`, `flipped`, `barsSinceFlip`, `distancePct` | A clean trend state with a trailing level; `flipped: true` is an event, `barsSinceFlip` says how mature the trend is |
+| `ichimoku` | `tenkan`, `kijun`, `tkCross`, `tkAboveKijun`, `cloud {spanA, spanB, top, bottom, color}`, `futureCloud {spanA, spanB, color}`, `pricePosition` (above_cloud / in_cloud / below_cloud), `chikouAbovePrice`, `signal` (bullish / bearish / neutral) | Price vs cloud is the trend, `futureCloud.color` is where the cloud is heading, TK cross is the trigger; `signal` is bullish only when all three agree |
+| `keltner` | `periods {ema, atr, multiplier}`, `upper`, `middle`, `lower` | Volatility channel (EMA ± ATR × multiplier); Bollinger inside it is the TTM squeeze |
+| `pivots` | `basis: previous_day`, `source {openTime, high, low, close}`, `pivot`, `r1…r3`, `s1…s3`, `priceAbovePivot`, `nearestResistance`, `nearestSupport` | Classic floor pivots from yesterday's daily candle regardless of the requested timeframe |
+| `levels` | `resistance[]` / `support[]` `{price, touches, swingHighs, swingLows, barsSinceLastTouch, distancePct}`, `range {high, low, positionPct}`, `fibonacci {leg, legStart, legEnd, retracedPct, levels[], nearestAbove, nearestBelow}`, `volumeProfile {poc, pocVolumePct, valueAreaHigh, valueAreaLow, pricePosition, nodes[]}` | More `touches` = more respected level. `range.positionPct` says where price sits in its 300-bar range. The Fibonacci leg is the dominant swing of the lookback; `retracedPct` is how much of it has been given back |
+
+## `summary` fields
+
+- `bias`: bullish / bearish / neutral; `score`: signed vote sum (|score| ≥ 30 sets a bias).
+- `trend`: `direction` (up / down / sideways), `strength` (absent / weak / moderate / strong),
+  `adx`, `emaStack`, `priceVsEma200`.
+- `momentum`: `rsi`, `rsiZone`, `macdHistogram`, `macdState`.
+- `volatility`: `atrPct`, `atrPercentile` (rank of today's ATR% in the last 100 bars — 90 means
+  unusually volatile), `bollingerBandwidthPct`, `state` (compressed / normal / expanded),
+  `squeeze`, `squeezeBars`.
+- `volume.ratioVsAverage`: current bar ÷ 20-bar average (the current bar is partial on the live
+  candle — a ratio of 0.2 early in a 4h bar is normal).
+- `patterns[]`: `{name, bias, barsAgo}` from 18 candlestick finders (engulfing, hammer, doji
+  variants, morning/evening star…). Patterns add notes, not score.
+- `notes[]`: quotable sentences; use them.
+
+## Timeframe defaults that read well
+
+- Intraday question → `["15m","1h","4h"]`
+- "How does X look" with no horizon → `["1h","4h","1d"]`
+- Position / investor framing → `["4h","1d"]` plus `levels` and `pivots`

package/bundled-skills/traderspy-trading-signals/SKILL.md

@@ -0,0 +1,156 @@
+---
+name: traderspy-trading-signals
+description: "Fetch and explain TraderSpy's AI crypto futures signals: entry, take-profit ladder, stop, triggers, status against the live price, and how recent signals resolved. Use for \"latest signals\"."
+category: finance
+risk: safe
+source: "https://github.com/target1m/traderspy-mcp/tree/069aae5a84640d671f92705a2e3bc0b62efca1e0/skills/trading-signals"
+source_repo: target1m/traderspy-mcp
+source_type: official
+date_added: "2026-09-25"
+author: target1m
+tags: [traderspy, crypto, trading-signals, futures, mcp]
+tools: [claude, cursor, gemini]
+license: MIT
+license_source: "https://github.com/target1m/traderspy-mcp/blob/069aae5a84640d671f92705a2e3bc0b62efca1e0/LICENSE"
+---
+
+# TraderSpy Trading Signals
+
+TraderSpy's AI signals are automated reads of the market: a named preset fires when several
+technical conditions line up on one crypto futures pair, a validator scores the alignment, and
+the published signal carries an entry, a take-profit ladder, a stop and — later — what actually
+happened. The signals feed at https://traderspy.app/signals publishes every one with its resolved
+outcome. Your job is to fetch,
+translate and contextualise them; the decision stays with the user.
+
+## When to Use
+
+- Use when the user mentions crypto signals, AI alerts, "any setups", "what is the AI seeing", long or short ideas, or TraderSpy alerts, even without the word "signal".
+- Use to explain one signal's entry, take-profit and stop levels and the conditions that triggered it, or whether it still stands at the current price.
+- Use when the user asks how signals have resolved (hits vs stops) over the last 4h to 7d.
+- Not for placing trades (no tool can), not for top-trader positioning (`@traderspy-smart-money`), not for indicator reads on a coin (`@traderspy-technical-analysis`).
+
+## Tools
+
+| Tool | Use it for | Key arguments |
+| --- | --- | --- |
+| `get_signals` | The list: newest first | `limit` 1–50 (default 20), `skip`, `importance` high / medium / low / all, `coin` |
+| `get_signal_details` | One signal in full, plus the live price | `signalId` (the `id` from the list) |
+| `get_signal_stats` | Aggregate hit rate over a window | `period` 4h / 8h / 24h / 7d |
+
+Three argument behaviours that are easy to get wrong:
+
+- **`importance` is inclusive downward.** `high` returns high only, `medium` returns high AND
+  medium, `low` returns everything. To show "medium and above" pass `medium`, not `low`.
+- **`coin` is a prefix match** on the pair name. `BTC` matches `BTCUSDT` and also `BTCDOMUSDT`;
+  check the `coin` field of each row before presenting "BTC signals".
+- **`limit` is what gets rendered.** In hosts that draw signal cards, every fetched row becomes a
+  card. If the user asks for "the last 5", pass `limit: 5` — do not fetch 20 and show 5.
+
+## What a signal row contains
+
+`strategyName` (the preset, e.g. "OBV Divergence Buy (4H)") · `action` buy / sell · `timeframe`
+1h / 4h / 1d · `price` at trigger · `targets[]` as `{label, type, pct}` where `pct` is measured from
+`price` · `triggeredConditions[]` in plain language · `importance` · `signalStrength` weak /
+moderate / strong / very_strong · `resolutionStatus` · `createdAt`.
+
+Convert target percentages to prices before showing them — users think in prices:
+
+- buy: TP = price × (1 + pct/100), SL = price × (1 − pct/100)
+- sell: TP = price × (1 − pct/100), SL = price × (1 + pct/100)
+- reward-to-risk at TP1 = TP1 pct ÷ SL pct (a 0.7 R:R is normal for this system, whose first target
+  sits close to entry; read it together with the hit rate from `get_signal_stats`, and do not call a
+  sub-1 R:R "bad" on its own).
+
+`resolutionStatus` is the outcome so far:
+
+| Status | Meaning |
+| --- | --- |
+| `pending` | Still inside its tracking window, nothing hit yet |
+| `tp1_hit` / `tp2_hit` / `tp3_hit` | Highest take-profit reached (a later stop touch does not downgrade it) |
+| `profit_locked` | Trend exhaustion detected before TP1; a partial gain was locked |
+| `stop` | Stop level touched before any take-profit |
+| `expired` | Window ended with neither side touched |
+
+Outcomes are judged on 1-minute wicks, so "TP1 hit" means price actually traded there. The
+tracking window is 24 × the signal timeframe (24h for a 1h signal, 4 days for a 4h signal); a 1h
+signal that is still `pending` two days later is stale rather than alive.
+
+## Workflows
+
+**"Latest signals" / "any setups?"** → `get_signals` with the limit the user implies (default 10 if
+they gave no number). Lead with the newest, group by side if several, and make the resolved ones
+visible — a list that mixes three winners, two stops and five pending rows should say so.
+
+**"Signals for SOL"** → `get_signals` with `coin`. If the result is empty, say there is no recent
+signal on that pair rather than offering a different pair as if it were the same thing.
+
+**"Is this signal still valid?" / "Should I still care about it?"** → `get_signal_details`, then
+compare `livePrice` with the entry, TP1 and SL in the same units:
+
+- distance from entry in % (signed the way the trade wants it: for a buy, positive = in profit)
+- whether a level has already been crossed (a `pending` buy with `livePrice` below the SL price
+  is finished in everything but paperwork — say so)
+- `history.highestPrice` / `lowestPrice` show the best and worst it has seen since entry
+- `indicatorValues` are the readings at trigger time, not now; if the user wants the current
+  picture, hand off to technical-analysis rather than re-reading stale values as if they were live
+
+**"How are the signals doing?"** → `get_signal_stats`. Explain the number honestly: `winRate` is
+hits ÷ (hits + stops) over signals CREATED in the window, so it ignores `pending` rows, and a 4h
+window is a handful of signals. Prefer `7d` for a track-record question and say how many signals
+it rests on (`total`, `pending`).
+
+## Presenting
+
+For a list, ALWAYS use a compact table and keep one signal per row:
+
+| Coin | Side | TF | Entry | TP1 | SL | Preset | Status | Age |
+
+For one signal, use this order: headline (coin, side, timeframe, preset, importance) → the levels
+as prices with the % in brackets → what triggered it (the `triggeredConditions`, lightly
+rephrased) → where price is now versus entry / TP1 / SL → the outcome so far → the one-line risk
+note. Quote `livePrice` with its timestamp when the answer depends on it.
+
+Do not invent an "AI review" if `aiReview` is null — most signals do not carry one. When it exists,
+report `score` and `decision` as the reviewer's opinion, not as a verdict.
+
+## Conduct
+
+- These are indicator alignments with published outcomes, not instructions. Report what the
+  signal says, what has happened to it and what has happened to signals like it; if the user asks
+  whether to take it, lay out what supports and what undercuts it and hand the decision back. Never
+  tell the user to buy, sell, size or leverage.
+- Historical hit rates describe the sample they were computed on. Never present a win rate as a
+  forecast or imply any outcome is assured.
+- Nothing in this connector trades. There is no order, close, transfer or withdrawal tool, by
+  design. If asked to execute, say so plainly; the decision and the trade stay with the user.
+- Every number comes from a tool result; if a field is null, say it is unavailable rather than
+  estimating it.
+- When the answer is about a specific trade idea, end with one plain sentence that crypto
+  derivatives are high-risk and this is market information, not financial advice. Once per answer
+  is enough.
+
+For the full field glossary, including the `indicatorValues` keys, read
+`references/signal-fields.md`.
+
+## Examples
+
+```text
+Show me the latest AI signals.
+Any signals on SOL right now?
+Is the latest BTC signal still valid? Show me where price is against its entry, targets and stop.
+How have the signals resolved over the last 7 days?
+```
+
+## Limitations
+
+- Needs the hosted TraderSpy MCP server connected in the client (`https://mcp.traderspy.app/mcp`, Streamable HTTP), authorized with OAuth or a personal key from https://traderspy.app/mcp. A free TraderSpy account is enough. Without the server the skill has no data to work from.
+- Tool calls are metered per day: 300 on the free tier, 5,000 on premium.
+- Every tool is read-only. Nothing here places, closes or modifies an order, and there is no withdrawal or transfer tool.
+- The output is market data and analysis for the user's own research, not investment advice.
+- Covers TraderSpy's own AI signals only. Signal statistics describe signals that already resolved; they are not a forecast.
+
+## Related Skills
+
+- `@traderspy-technical-analysis` (called `technical-analysis` in the text above, its upstream ID) - the current chart read on the signal's coin
+- `@traderspy-market-briefing` - the freshest signals inside a market overview

package/bundled-skills/traderspy-trading-signals/references/signal-fields.md

@@ -0,0 +1,64 @@
+# Signal field glossary
+
+## `get_signals` → `data[]`
+
+| Field | Type | Notes |
+| --- | --- | --- |
+| `id` | string | Pass to `get_signal_details` |
+| `strategyName` | string | Preset display name. The timeframe is usually in the name, e.g. "(4H)" |
+| `importance` | high / medium / low | Preset tier, not a per-signal confidence |
+| `action` | buy / sell | Direction |
+| `signalStrength` | weak / moderate / strong / very_strong | How many optional conditions matched; capped at `strong` unless the entry was fresh |
+| `coin` | string | Binance Futures pair, e.g. `WCTUSDT` |
+| `timeframe` | 1h / 4h / 1d | Chart timeframe the preset evaluated |
+| `price` | number | Price when the signal was published; targets are measured from it |
+| `targets[]` | `{label, type, pct}` | `type` is `tp1` / `tp2` / `tp3` / `sl`; `pct` is unsigned distance from `price` |
+| `triggeredConditions[]` | string | Plain-language conditions that fired, e.g. "Volume 2.2x average" |
+| `resolutionStatus` | see table in SKILL.md | `pending` when nothing has resolved yet |
+| `createdAt` | ISO date | Publication time |
+
+`pagination` = `{ limit, skip, total, hasMore }`. `total` is the count matching the filter — it is
+the size of the whole public archive when no filter is given, not "signals today".
+
+## `get_signal_details` adds
+
+| Field | Notes |
+| --- | --- |
+| `livePrice` | Current price of the pair at call time — the number that decides whether the idea still stands |
+| `indicatorValues` | Readings captured at trigger time. Common keys: `rsi`, `macd_histogram`, `adx`, `obv_slope`, `obv_flow_pct`, `volume_ratio`, `williams_r`, plus `validation_*` keys written by the publisher's scoring pass (`validation_score` 0–100, `validation_confidence`, `validation_rsi`, `validation_atr_pct`, `validation_btc_trend_1d`, `validation_btc_rsi_4h`, `validation_directional_move_pct` = how far price had already moved in the trade's direction when it was published) |
+| `aiReview` | `{ score, decision, analysis }` or null. Null is the normal case |
+| `history.entryPrice` | Price at entry (equals `price` unless recalibrated) |
+| `history.currentPrice` | Last tracked price; may be null on a fresh signal |
+| `history.highestPrice` / `lowestPrice` | Extremes seen since entry on 1-minute data |
+| `history.resolution` | `{ status, price, time }` once resolved; `price` is the level that was hit, so realised move = (resolution.price − entryPrice) / entryPrice, sign-flipped for sells |
+
+Interpretation notes:
+
+- `validation_score` around 70–80 has historically been the sweet spot; the very highest scores
+  tended to be late entries. Do not rank signals by score alone.
+- `validation_directional_move_pct` above ~1.5% means the move was already under way at
+  publication — worth mentioning as "the entry chased the move".
+- The published TP2/TP3 are a ladder (≈1.7× and 2.4× TP1); they exist so a winner can be
+  graded beyond TP1, not as three independent forecasts.
+
+## `get_signal_stats`
+
+`{ total, targetHits, stopped, pending, highCount, winRate }` over signals created in the period.
+
+- `targetHits` counts `tp1_hit`, `tp2_hit`, `tp3_hit` and `profit_locked`.
+- `pending` = total − hits − stops (so it also contains `expired` rows).
+- `winRate` = round(hits ÷ (hits + stops) × 100), null when nothing has resolved.
+- `highCount` = how many were high importance.
+
+The 4h and 8h windows are small samples; treat single-digit `total` as noise.
+
+## Resolution mechanics worth knowing
+
+- Tracking window = 1×, 4× and 24× the signal timeframe (1h → 1h/4h/1d; 4h → 4h/16h/4d).
+- TP and SL are judged on 1-minute highs/lows starting the minute after entry; on a same-minute
+  ambiguity the stop is recorded first (conservative).
+- TP2 and TP3 count only after TP1; after TP1 a later stop touch does not downgrade the result.
+- `profit_locked` is set when a trend-exhaustion check fires before TP1 and locks a partial gain.
+- Only a subset of signals is ever marked `expired`; a signal that finishes its window untouched
+  can stay `pending` indefinitely. Treat old `pending` rows as roughly flat outcomes, not as open
+  ideas.

package/bundled-skills/unified-ai-gateway/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: unified-ai-gateway
-description: Operate and evaluate Unified AI System through nine governed MCP tools, including provider-free prompt enhancement, while preserving fake-provider, authorization, and evidence boundaries.
+description: Operate and evaluate Unified AI System through fifteen governed MCP tools, including provider-free prompt enhancement, while preserving fake-provider, authorization, and evidence boundaries.
 category: ai-ml
 risk: critical
 source: https://github.com/happy520ai/unified-ai-system/tree/master/skills/unified-ai-gateway
@@ -35,7 +35,7 @@ review is required before changing this pinned procedure.
 ## Prerequisites And Setup
 
 1. Confirm that Codex CLI and Docker are installed and Docker is running.
-2. If the nine tools are already visible, skip setup and do not register a
+2. If the 15 tools are already visible, skip setup and do not register a
    duplicate server.
 3. Explain the first stage: it downloads one reviewed platform from the
    immutable `0.4.9` multi-platform index into Docker's cache, inspects its
@@ -133,7 +133,7 @@ codex mcp get unified-ai-system --json
 ```
 
 8. Restart Codex or open a new task, then use `/mcp verbose` to confirm that all
-   nine tools are available. Remove the registration when it is no longer
+   15 tools are available. Remove the registration when it is no longer
    wanted:
 
 ```bash

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "opencode-skills-collection",
-  "version": "4.0.73",
+  "version": "4.0.75",
   "description": "OpenCode CLI plugin that automatically downloads and keeps skills up to date.",
   "main": "dist/index.js",
   "types": "dist/index.d.ts",