Equity Research Snapshot
Turn consensus estimates, fundamentals, price history, and macro data into a structured stock research view.
The skill limits itself to named research-data MCP calls and drafting an analysis snapshot, with no stated write, trading, or other external side effects. It does not specify least privilege, user confirmation, data flows, sensitive-data handling, rollback, or per-claim source disclosure, so points are deducted.
The ordered workflow, tool roles, and output structure are reasonably coherent. However, there are no tool schemas, dependency details, abnormal-input handling, retry behavior, or diagnosable failure messages; static calibration also limits the score.
The audience, equity-research scenarios, trigger wording, and expected data categories are fairly clear. Non-fit boundaries, Chinese-language support, mainland-China network reachability, and boundaries around valuation and recommendations are not stated, so adaptability is only partial.
The document provides headings, a tool inventory, numbered workflow, table templates, and output requirements. Repository context supplies Apache-2.0 licensing, verified organizational provenance, CI, and contribution guidance, but the skill itself lacks installation notes, versioning, changelog, FAQs, maintenance ownership, and known limitations.
The workflow could produce a useful snapshot covering consensus, financials, prices, macro context, and an investment thesis. Yet forward P/E, EV/EBITDA, peer/history comparisons, fair value, and buy/hold/sell conclusions lack precise data definitions and validation rules, leaving substantial professional review; static evidence supports no more than a mid-level score.
Named data tools and explicit calculation targets provide limited auditability. There are no committed tests, representative outputs, third-party execution records, source-to-claim mappings, or cross-validation procedures; the repository CI validates plugin structure rather than this skill's key paths.
- The skill requests buy/hold/sell recommendations and a fair-value range without defining compliance review, user confirmation, jurisdictional safeguards, or disclaimer handling; outputs must not be treated as investment advice.
- Actual MCP availability, authentication, field semantics, data freshness, and mainland-China network reachability are not established by the skill file.
- Valuation and return calculations lack specified currency, accounting basis, as-of date, debt definitions, outlier handling, and citation rules; professional review is required.
What does this skill do, and when should you use it?
This Agent Skill produces comprehensive equity research snapshots. It uses MCP tools to retrieve IBES analyst consensus, company financials, historical equity prices, and macroeconomic indicators. The workflow connects each data point to an investment thesis and asks where consensus may be wrong. Its specified output covers consensus estimates, financial trends, valuation context, and an investment thesis.
It calls qa_ibes_consensus for FY1 and FY2 EPS, revenue, EBITDA, and DPS estimates, analyst counts, and dispersion; qa_company_fundamentals for three to five years of income statement, balance sheet, and cash flow data; qa_historical_equity_price for one-year prices, returns, 52-week range position, and beta; tscc_historical_pricing_summaries for three months of recent pricing and volume detail; and qa_macroeconomic for indicators such as GDP, CPI, unemployment, and PMI. It then synthesizes the results into standardized research tables and an investment thesis.
- An equity analyst covering a company needs a concise view of consensus, historical financials, and price performance.
- A research professional compares companies by forecast expectations, operating trends, and valuation context.
- An investor evaluates whether a stock price already reflects prevailing market consensus.
- A research team builds an investment case that links company fundamentals with the macroeconomic backdrop.
- A portfolio manager reviews performance, forecast dispersion, and potential thesis implications for a holding.
What are this skill's strengths and limitations?
- Combines consensus, fundamentals, price history, and macroeconomic context in one workflow.
- Defines a clear sequence of tool calls and a standardized output structure.
- Explicitly focuses on identifying where market consensus may be wrong.
- Supports multiple consensus metrics, including EPS, revenue, EBITDA, and DPS.
- Depends on several MCP tools and their underlying data; no standalone fallback is documented.
- MCP data access may require a provider subscription or API key.
- The source provides no evidence of accuracy benchmarks, test coverage, or platform testing.
- The skill supplies the research workflow rather than the data, so outputs require professional review.
How do you install this skill?
The repository can be added in Claude Cowork through Settings → Plugins → Add plugin by pasting https://github.com/anthropics/financial-services, or by uploading a directory under plugins/. In Claude Code, the README documents claude plugin marketplace add anthropics/financial-services followed by installation of selected plugins. The source does not document a standalone installation command for the specific skill at plugins/partner-built/lseg/skills/equity-research/.
How do you use this skill?
In an environment with the relevant MCP connections configured, submit a focused request such as: “Create an equity research snapshot for a company covering FY1/FY2 consensus, three to five years of fundamentals, one-year price performance, three months of recent pricing detail, and the primary-market macro backdrop.” The skill follows its documented tool-chaining workflow and returns structured research content.