Update to Daloopa

{
  "name": "guidance-tracker",
  "description": "Track management guidance accuracy over time for a given company",
  "included_files": [
    {
      "relative_path": "agents/openai.yaml",
      "size_in_bytes": 241
    }
  ],
  "skill_md_contents": "---\nname: guidance-tracker\ndescription: Track management guidance accuracy over time for a given company\n---\n\nTrack management guidance accuracy for the company named in the user's request. If no ticker or company is provided, ask for one before proceeding.\n\n**Before starting, read `../data-access.md` for data access methods and `../design-system.md` for formatting conventions.** Follow the data access detection logic and design system throughout this skill.\n\nFollow these steps:\n\n## 1. Company Lookup\nLook up the company by ticker using `discover_companies`. Capture:\n- `company_id`\n- `latest_calendar_quarter` — anchor for all period calculations below (see `../data-access.md` Section 1.5)\n- `latest_fiscal_quarter`\n- Firm name for report attribution (default: \"Daloopa\") — see `../data-access.md` Section 4.5\n\n## 2. Discover Guidance Series\nSearch for series with keywords like \"guidance\", \"outlook\", \"estimate\", \"forecast\", \"target\" to find all available guidance metrics. Common guidance series include:\n\n**Financial guidance:**\n- Revenue guidance (quarterly and/or annual)\n- EPS guidance\n- Operating income / margin guidance\n- EBITDA guidance\n- Segment-level revenue guidance\n- CapEx guidance\n- Free Cash Flow guidance\n\n**Operational KPI guidance** — many companies guide on KPIs, and tracking these beats/misses is often more informative than financial guidance:\n- Subscriber / user count guidance (e.g., \"we expect to add X million subscribers\")\n- Unit shipment guidance (e.g., \"iPhone units\", \"deliveries\")\n- ARPU / ASP guidance\n- Same-store sales guidance\n- GMV / bookings guidance\n- Net revenue retention guidance\n- Store openings / closings guidance\n- Production volume / capacity guidance\n\nSearch explicitly for KPI-specific guidance series using terms like \"subscriber guidance\", \"unit guidance\", \"ARPU guidance\", \"same-store sales outlook\", \"deliveries forecast\", \"bookings target\". These are separate from financial guidance and often reside in different series.\n\n## 3. Pull Guidance Data\nCalculate 8+ quarters backward from `latest_calendar_quarter`. Pull all discovered guidance series for those periods.\n\n## 4. Pull Actual Results\nFor each guidance metric, pull the corresponding actual result series for the same periods.\n\n## 5. Build Guidance vs Actuals Tracker\nCRITICAL OFFSET RULES:\n- **Quarterly guidance**: Guidance from Q(N) earnings call applies to Q(N+1) results. Compare Q(N) guidance -> Q(N+1) actual.\n- **Annual guidance from Q1/Q2/Q3**: Applies to current fiscal year. Compare to FY actual.\n- **Annual guidance from Q4**: Applies to NEXT fiscal year. Compare to next FY actual.\n\nFor each guidance-actual pair, calculate:\n- Guidance value\n- Actual value\n- Delta (Actual - Guidance)\n- Beat/Miss % ((Actual - Guidance) / |Guidance| x 100)\n- Classification: Beat / In-line / Miss (use +/-1% threshold for in-line)\n\n## 6. Pattern Analysis\nAnalyze the guidance track record:\n- Overall beat rate (% of quarters where actual > guidance)\n- Average beat/miss magnitude\n- Trend in guidance accuracy (getting tighter? more conservative? less reliable?)\n- Any metrics where management is notably conservative or aggressive\n- Guidance range width trends (if range guidance is given)\n\n**Management credibility assessment:**\n- If the company consistently beats by a similar margin, call out sandbagging — this suggests management is deliberately setting low bars, which can mask underlying deceleration. A 100% beat rate is not necessarily bullish; it may mean guidance is uninformative.\n- If guidance has been cut or missed, assess whether management acknowledged the miss honestly or buried it in adjusted metrics.\n- Flag any pattern where qualitative language (\"strong demand,\" \"robust pipeline\") didn't translate to actual results.\n\n## 7. Commentary from Filings\nSearch SEC filings/documents across multiple queries to build a complete picture of guidance practices. If any search returns empty, try alternative keywords before giving up.\n\n- **Explicit guidance language**: Try \"guidance\", \"outlook\"; fallback to \"expect\", \"anticipate\", \"forecast\"\n- **Qualitative / directional guidance**: Try \"similar to\", \"consistent with\", \"growth rate\"; fallback to \"low single digit\", \"mid single digit\", \"high single digit\", \"double digit\", \"sequential\"\n  - Many companies provide directional revenue guidance on earnings calls (e.g., \"similar to the March quarter\" or \"low-to-mid-single-digit growth\") rather than numeric ranges. Capture these and compare against actual growth rates.\n- **Guidance methodology changes**: Try \"change\", \"methodology\", \"no longer providing\"; fallback to \"withdraw\", \"suspend\", \"discontinue\"\n  - Flag any quarters where the company changed what metrics it guides on, or withdrew guidance entirely\n- **Key drivers behind guidance**: Try \"assumes\", \"includes\", \"excludes\"; fallback to \"headwind\", \"tailwind\", \"impact\"\n  - Capture what management said about the assumptions underpinning their guidance (e.g., FX assumptions, macro assumptions, one-time items included/excluded)\n\nExtract direct management quotes where available and cite the document source.\n\n## 7.5. Guidance Read-Throughs to Adjacent Companies\n\nWhen a company raises, cuts, or materially changes its guidance, the implications often matter more for adjacent names than for the company itself. This section translates guidance signals into actionable read-throughs.\n\n**For each major guidance change identified in the tracker, analyze the implications for adjacent companies:**\n\n**Identify who is affected by this company's guidance:**\n- **Suppliers**: Revenue/CapEx guidance changes directly affect supplier order books. A CapEx guidance raise is a near-term purchase order for equipment/component suppliers. A revenue guide-down signals softer demand flowing upstream.\n- **Customers**: If this company supplies critical inputs, pricing or capacity guidance affects customer margins. Guiding for price increases = margin headwind for customers. Guiding for capacity expansion = supply relief.\n- **Competitors**: Guidance on market growth, pricing environment, or demand trends is often the most honest signal about the competitive landscape. If Company A guides for share gains, that's a direct share loss for Company B.\n- **Channel partners / distributors**: Volume guidance changes affect channel inventory and distributor revenue.\n\n**For each read-through (aim for 4-6), state:**\n1. **The guidance data point** — which metric changed, by how much, and in which quarter's call\n2. **The affected company** (ticker + name)\n3. **The implication** — bullish or bearish, with specific logic\n4. **Timing** — is this a next-quarter impact or a multi-quarter trend?\n\n**Focus on the highest-signal guidance changes:**\n- Guidance raises after a period of conservatism → strong signal that the underlying business is inflecting\n- Guidance cuts or \"reaffirmed\" when the market expected a raise → often more bearish than an explicit cut\n- New metrics being guided on (or old metrics withdrawn) → management is redirecting attention, which itself is a signal\n- Segment-level guidance changes → more specific read-throughs than consolidated figures\n- KPI guidance (subscriber adds, unit volumes, ARPU) → often the most direct read-through to suppliers and competitors\n\n**Example:**\n- \"NFLX raised Q2 subscriber guidance from +5M to +8M → **Negative for DIS+, WBD**: attention economy is zero-sum; NFLX's accelerating growth likely pressures competing streamers' subscriber adds. **Positive for cloud/CDN names (AMZN/AWS, NET)**: more streaming = more infrastructure demand.\"\n- \"TSMC raised full-year CapEx guidance by $4B (from $32B to $36B) → **Positive for ASML**: TSMC is ASML's largest customer; incremental CapEx skews toward EUV tools. **Positive for AMAT, LRCX, KLAC**: broader equipment spend benefits all semicap names.\"\n\n**Web research for validation:**\nRun 1 targeted search: `\"{TICKER} guidance change implications read through {year}\"` — analyst commentary on cross-company signals from guidance moves.\n\nPresent as a structured section in the report after the Pattern Analysis, grouped by guidance change (each major guide raise/cut gets its own sub-block with the read-throughs beneath it).\n\n## 8. Save Report\nSave to `reports/{TICKER}_guidance_tracker.html` using the HTML report template from `../design-system.md`. Write the full analysis as styled HTML with the design system CSS inlined. This is the final deliverable — no intermediate markdown step needed.\n\nThe report should include:\n- Summary header with company name, ticker, and period covered\n- Quarter mapping reference table showing the +1 offset explicitly:\n  ```\n  | Guidance Source Quarter | Guidance Applies To | Actual Result Quarter |\n  | CQ1 2024               | CQ2 2024            | CQ2 2024              |\n  | CQ2 2024               | CQ3 2024            | CQ3 2024              |\n  ```\n  This makes the offset rule visible and auditable for every row in the tracker.\n- Main tracker table with columns: Guidance Source, Metric, Guidance, Actual Period, Actual, Delta, Beat/Miss (with Daloopa citations on all values)\n- Summary statistics (beat rate, avg beat/miss by metric)\n- Pattern analysis narrative\n- Guidance read-throughs to adjacent companies (grouped by guidance change, with affected tickers and implications)\n- Key guidance quotes from filings with document citations\n\nAll financial figures must use Daloopa citation format: `<a href=\"https://daloopa.com/src/{fundamental_id}\">$X.XX million</a>`\n\nTell the user where the HTML report was saved.\n\nHighlight the key patterns (e.g., \"Management has beat revenue guidance 7 of the last 8 quarters by an average of 2.3%\"). Include an honest credibility verdict: Is management's guidance informative or performative? Should investors trust the forward guidance, and if not, what should they anchor to instead?\n"
}