What does this skill do, and when should you use it?
This skill produces 8–12-page, 3,000–5,000-word earnings update reports for companies already under coverage. It focuses on beat-or-miss analysis, key metrics, guidance changes, revised forward estimates, valuation, and the effect on the investment thesis. The specified format includes 1–3 summary tables, 8–12 charts, and clickable citations for figures, tables, and key statistics. The primary deliverable is a DOCX report, with an optional XLS model update.
Collects the latest earnings release, Form 10-Q, earnings call transcript, investor materials, consensus estimates, and prior guidance; checks the current date, release date, and transcript date before analysis; evaluates metric variances, segment performance, margins, guidance, estimate revisions, and thesis implications; uses Python with matplotlib, pandas, and seaborn for 8–12 charts; and creates a DOCX report containing an earnings summary, results analysis, guidance, thesis assessment, valuation, updated estimates, and a clickable sources section.
- An equity analyst covering a company who needs a report within 24–48 hours of its quarterly earnings release.
- A research team explaining why revenue, EPS, or operating metrics beat or missed market expectations.
- An analyst updating forward estimates and the investment thesis after new quarterly data.
- An institutional research team preparing a concise update for clients already familiar with the company.
How do you install this skill?
- The workflow depends on web search and several overseas data sources but provides no operational fallback for offline, restricted-network, or unavailable-source conditions.
- The DOCX skill, XLS skill, and chart-generation environment are not defined within the assessed skill; humans should verify files, numbers, citations, hyperlinks, and estimate calculations before delivery.
- Financial research may involve sensitive or regulated information, yet the skill does not specify data minimization, access control, retention, user confirmation, or publication approval.
- The output specification is primarily an English institutional-research format and does not establish Chinese terminology, Chinese layout, or mainland-China network usability.
- Network access
- Local filesystem
PythonmatplotlibpandasseabornDOCX skill
The README says to install the collection in Cowork through Settings → Plugins → Add plugin by pasting https://github.com/anthropics/financial-services and selecting the relevant agent or vertical. For Claude Code, it documents: claude plugin marketplace add anthropics/financial-services, followed by installation of the relevant agent or the equity-research vertical plugin. The supplied material does not document a standalone installation command for this individual skill.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/anthropics/financial-services.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/plugins/agent-plugins/earnings-reviewer/skills/earnings-analysis" ~/.claude/skills/
rm -rf "$tmp"Generated from the source repository and skill path; it copies only this skill's folder. If the author's install steps above differ, follow those first. To scope it to one project, replace ~/.claude/skills with that project's .claude/skills.
How do you use this skill?
Once installed, send your agent any of these to trigger it:
- Create an earnings update for [Company] Q3 2024
- Analyze [Company]'s quarterly results
- Post-earnings report for [Company]
In an environment where the skill is installed and callable, use a prompt such as "Create an earnings update for [Company] Q3 2024", "Analyze [Company]'s quarterly results", or "Post-earnings report for [Company]". The company should already be under coverage; the workflow requires searching for the latest earnings, confirming the release is within the last three months, and checking that the transcript date matches the release date.
What are this skill's strengths and limitations?
- Provides clear output constraints for length, charts, tables, structure, and DOCX delivery.
- Prioritizes beats or misses, estimate revisions, and thesis impact, making it suited to fast post-earnings work.
- Requires specific sources and clickable links for figures, tables, and key data.
- Defines a five-phase workflow covering collection, analysis, charting, report creation, and quality control.
- Limited to companies already under coverage and is not intended for initiation reports.
- Not intended for flash notes or quick takes.
- Requires Python charting libraries and a DOCX skill; no test suite or platform validation evidence is provided.
- Requires web searching for current materials, but the source does not provide data-provider access.
- The source specifies ratings and price targets in the report format but does not explain how to independently validate them.
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| Earnings Update Researcher this page ✓ Anthropic · Official | 54 · Use with care | ★ 40k | 1mo ago | Apache-2.0 |
| Equity Research Initiation Workflow ✓ Anthropic · Official | 44 · Not recommended | ★ 40k | 1mo ago | Apache-2.0 |
| Financial Model Refresh ✓ Anthropic · Official | 47 · Use with care | ★ 40k | 1mo ago | Apache-2.0 |
| 3-Statements-Ultra | 55 · Use with care | ★ 85 | 5mo ago | — |
| Earnings Team: Four Masters Parallel Analysis + WeChat Publishing | 49 · Use with care | ★ 17k | 13d ago | MIT |
Compared with an initiation report, this skill is shorter, faster, and focused on new quarterly information, with fewer tables and figures. Compared with a flash note or quick take, it requires a fuller 8–12-page report, 8–12 charts, updated estimates, and a complete sources section.
How did FollowSkills review this skill?
The skill requires specific source attribution and clickable links for figures, tables, and key data, and scopes the work to quarterly updates for companies already under coverage. It requests no trading execution or destructive action. However, it relies on web search, SEC filings, third-party consensus data, and external services without specifying least-privilege permissions, sensitive financial-data handling, user confirmation, rollback, or link-safety validation, so points are deducted.
The workflow, checklists, and date-mismatch rules provide useful guidance for the happy path and stale or inconsistent materials. Nevertheless, it depends on an undefined DOCX skill, web search, and potentially unavailable data sources; there are no skill-specific tests, reproduced outputs, or clear tool-failure handling. The score remains within the static-review ceiling and is reduced accordingly.
Trigger phrases, target scenarios, and non-fit cases are relatively clear: the skill targets post-earnings updates for companies already covered and specifies report contents. Boundaries are incomplete when transcripts, consensus data, or prior coverage materials are unavailable. The templates assume English-language institutional research and Word output, do not declare Chinese support or mainland-China reachability, and rely on overseas web sources such as SEC, so points are deducted.
The documentation has layered sections, phased workflow, citation rules, templates, quality checklists, and stable output naming. Repository evidence supplies an Apache-2.0 license, official provenance, and CI validation signals. The skill itself lacks a version, changelog, named maintenance owner, update path, dependency-install instructions, and FAQ; the referenced DOCX/XLS skills are also not concretely identified, so points are deducted.
The goal, report structure, estimate-update process, chart requirements, and delivery format are concrete and could support a standardized quarterly update. Static files cannot establish that the system actually creates a usable DOCX, embeds charts, calculates estimate changes correctly, or validates hyperlinks. Professional review remains necessary, so the score is capped by static calibration and reduced for the lack of execution evidence.
The skill specifies source, date, quarter-consistency, consensus-data, and hyperlink checks, while repository CI provides plugin-validation and secret-scan evidence. There are no skill-specific tests, real execution artifacts, or independent third-party reviews covering the key path; static instructions alone cannot verify compliance, so points are deducted.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →