Earnings Preview Analyst
Frame pre-earnings expectations, scenarios, and likely stock reactions.
The skill mainly requests public-market research and drafting, with no instruction to trade, modify records, or cause other external effects, so inherent permission risk is limited. It does not require user confirmation, disclose concrete data flows or service dependencies, or specify sensitive-data handling, recovery, or investment-research disclaimers, so points are deducted.
The workflow is internally coherent and covers company, quarter, estimates, scenarios, catalysts, and output structure. It lacks abnormal-input handling, missing-data procedures, search-failure feedback, dependency availability guidance, and reproducible tests. The static-review ceiling applies, so the score remains below 10.
The name, description, trigger phrases, and target scenario are clear, and sector-specific metrics are listed. However, non-fit boundaries, input requirements, market coverage, Chinese-language support, mainland-China reachability, and behavior when the company or quarter is missing are unspecified, so points are deducted.
The documentation is readable and progressively organized around steps, tables, and an output checklist. The repository supplies Apache-2.0 licensing, official-organization provenance, and CI maintenance signals, but the skill lacks versioning, changelog, FAQ, explicit ownership, installation/dependency notes, and known limitations, so points are deducted.
The skill can produce the intended earnings-preview framework, scenario table, and catalyst checklist with practical analytical value. However, it depends on external retrieval and does not define estimate conventions, source-quality rules, price-reaction calculation methods, or usable fallback outputs when data is unavailable; substantial professional review remains necessary, so points are deducted.
The document asks the user to record estimate sources and dates and points to historical reactions and options-implied moves as research inputs. It contains no skill-specific tests, representative outputs, independent reproduction material, or third-party execution evidence. Under static calibration, the score cannot exceed 5.
- Consensus estimates, whisper numbers, earnings timing, historical price reactions, and options-implied moves may be missing, stale, or definitionally inconsistent; verify source, date, timezone, and methodology for each item.
- The skill requests stock-reaction analysis and a trading setup but does not clearly frame the result as a research draft or specify investment-advice, compliance-review, and user-confirmation boundaries.
- No fallback behavior is defined for unavailable search, paid-data access, inapplicable sector metrics, or an unspecified company or quarter.
What does this skill do, and when should you use it?
This skill prepares research before a company reports quarterly earnings. It gathers consensus estimates, the reporting schedule, and prior-quarter management commentary, then turns them into a company-specific watchlist. It builds bull, base, and bear cases with operational drivers and possible stock reactions. The deliverable is a one-page earnings preview for expectation setting and trading preparation.
Identifies the company and reporting quarter; uses web search to gather consensus revenue, EPS, and segment estimates; finds the earnings date and whether the release is pre-market or after-hours; reviews the prior-quarter earnings call for guidance and commentary; organizes financial and sector-specific operating metrics; builds bull, base, and bear scenarios; lists three to five catalysts likely to determine the stock reaction; and produces a one-page preview including recent performance and the options-implied move.
- An equity researcher preparing a ranked watchlist before a company’s quarterly earnings release.
- A portfolio manager comparing published consensus with potential bull, base, and bear outcomes.
- A sector analyst selecting the most relevant operating metrics for a technology, retail, industrial, financial, or healthcare company.
- A trader comparing the market’s options-implied move with scenario-based expectations before earnings.
What are this skill's strengths and limitations?
- Covers the core pre-earnings workflow from estimate gathering through scenarios and catalysts.
- Combines financial metrics with sector operating metrics, historical earnings reactions, and options-implied expectations.
- Requires the source and date of consensus estimates to be noted, which is useful because estimates change.
- Depends on web search and external data; the SKILL.md names no specific provider, quality standard, or automated data connector.
- Does not supply company-specific forecast values automatically; revenue, EPS, and price-reaction assumptions must be researched and populated.
- The source provides no test suite, standalone installation flow, or platform validation record.
How do you install this skill?
The source does not document a standalone installation command for this skill. In Cowork, open Settings → Plugins → Add plugin, paste https://github.com/anthropics/financial-services, and select the relevant agent or vertical plugin; alternatively, zip plugins/agent-plugins/earnings-reviewer/ and upload it. The README also documents Claude Code marketplace installation for the collection, but not a dedicated earnings-preview install command.
How do you use this skill?
Before a company reports, enter “earnings preview” or “what to watch for [company] earnings.” Other supported trigger phrases include “pre-earnings,” “earnings setup,” and “preview Q[X] for [company].” The expected output includes the company, quarter, earnings date, consensus table, ranked metrics to watch, bull/base/bear scenarios, catalyst checklist, and trading setup.