Earnings preview
Skill yeshelloAB/investment-analysis-skills/skills/earnings-preview
Pre-earnings setup for any listed company ahead of a quarterly, half-year, or full-year result. Builds a structured decision framework: consensus estimates, bull/base/bear scenarios, key metrics to watch, thesis pillar checklist, and historical stock reaction context. Pairs with earnings-analysis (post-print verdict) and thesis-tracker (pillar updates). Covers US (NYSE/NASDAQ) and AU (ASX) markets. Triggers on: "earnings preview", "what should I watch for [company] earnings", "pre-earnings", "earnings setup", "preview Q[X] for [company]", "prepare me for [company] results", "what's consensus for [company]", "[company] reports [date], what do I need to know", "earnings preview for [ticker]", or any request to prepare ahead of an upcoming earnings release.From its SKILL.md
npx -y skills add yeshelloAB/investment-analysis-skills --skill earnings-previewAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- skips confirmationTells the agent to proceed without asking first, 1 time: "Then proceed. Do not ask for anything else.".
- 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
6.9 KB, ~1.5k tokens by cl100k_base, as published. Nobody here has run it
Earnings Preview
Role
You prepare the investor to enter an earnings event with a structured position: what consensus expects, what the thesis requires, what scenarios are plausible, and what specific metrics will determine whether conviction rises or falls after the print.
This is a decision framework built before the result, not a prediction, and not a research report. Its job is to eliminate noise so the investor can assess the actual print quickly when it drops.
The output from this skill feeds directly into earnings-analysis. When running the post-print review, reference this preview document.
Step 0: Calibrate Depth
| Mode | Trigger | Output |
|---|---|---|
| Quick | "just the key numbers", "brief", "what's consensus" | Consensus table + 3 key metrics |
| Full (default) | No modifier, or any mention of thesis / positioning | All steps |
Step 1: Confirm Inputs
- Company and ticker (include exchange suffix for ASX: e.g. WDS.AX)
- Reporting period: Q1/Q2/Q3/Q4 or H1/H2/FY and fiscal year
- Earnings date and time: confirm via web search. For AU: note pre-market (ASX open) or post-close (after 4pm AEST)
- Existing thesis: check Theses folder for
[TICKER]-thesis.md. Load if found. - Position: held (long), watchlist, or none
Then proceed. Do not ask for anything else.
Step 2: Consensus Snapshot
Pull via web search. For US: Yahoo Finance, Seeking Alpha, company IR. For ASX: CommSec, ASX announcements, company IR. Note analyst count: thin coverage (1-3 analysts) means higher surprise potential and a wider confidence interval.
Company: [Name] ([TICKER]) Reporting period: [Q/H/FY] [Year] Earnings date: [Day DD Mon YYYY, Time] Analysts covering: [X]
| Metric | Consensus | Prior Period Actual | YoY Change | Estimate Range (High / Low) |
|---|---|---|---|---|
| Revenue | ||||
| EBITDA / Operating Profit | ||||
| NPAT / EPS | ||||
| Gross Margin | ||||
| [Key sector metric 1] | ||||
| [Key sector metric 2] | ||||
| Forward guidance | N/A (wait for result) | : | : | : |
A wide estimate range signals high uncertainty: flag this as it implies a larger potential price move.
Step 3: Key Metrics Framework
Identify the 3-5 metrics that will actually move this stock on results day: often not the headline figures. State why each matters right now.
For each metric:
- Metric: [Name and expected value or direction]
- Why it matters this print: [1 sentence]
- Threshold: [What level / direction is positive vs. concerning]
Common sector-specific metrics:
- Technology / SaaS: ARR growth, net revenue retention, RPO, operating leverage
- Resources / Mining (ASX): Realised prices vs. spot, production volumes vs. guidance, AISC, capex vs. budget
- Banks / Financials: NIM direction, credit quality, loan growth, cost-to-income
- Healthcare: Script volumes, pipeline updates, reimbursement decisions
- Industrials: Backlog, book-to-bill, price vs. volume mix
- Retail: Same-store sales, gross margin, inventory levels
Step 4: Scenario Framework
| Scenario | Probability | Revenue vs. Consensus | EPS vs. Consensus | Key Driver | Expected Stock Reaction |
|---|---|---|---|---|---|
| Bull | [%] | [+X%] | [+X%] | [What drives it] | [+X% to +X%] |
| Base | [%] | [~flat] | [~flat] | [In-line, guidance intact] | [Range] |
| Bear | [%] | [-X%] | [-X%] | [What drives it] | [-X% to -X%] |
Probabilities must sum to 100%. For each scenario, name the confirmation signal: the single piece of evidence in the result or commentary that immediately tells you which scenario has played out.
Step 5: Historical Reaction Context
How has this stock moved on the last 3-4 prints?
| Period | Beat / Miss / In-line | Revenue | EPS | Guidance | Stock Day-1 Move |
|---|---|---|---|---|---|
| [Q/H/FY Year] |
Note any pattern: does this stock reward beats, or is it guidance-dependent? Source via web search.
Step 6: Thesis Pillar Checklist (only if thesis file exists)
| Pillar | What the Print Will Show | Confirmation Signal | Concern Signal |
|---|---|---|---|
| [Pillar name] | [Specific metric or disclosure] | [What confirms it's intact] | [What challenges it] |
Work through this table live when the result drops: note ✅/⚠️/❌ next to each pillar. Then hand these notes to earnings-analysis for the post-print verdict.
If no thesis file exists, flag this and suggest running thesis-tracker after the result.
Step 7: Output
Quick mode: Consensus table + three key metrics. Inline only, no file.
Full mode:
- Deliver inline: lead with earnings date and key metrics (most time-sensitive), then scenarios, then pillar checklist.
- Save to the user's workspace folder:
Stock Strategy/Stock Investment Analysis/Earnings-Previews/[TICKER]-[Period]-preview.md(or equivalent) - Confirm file saved and path.
- Note: when the result drops, run earnings-analysis and reference this preview file.
Important Notes
- For ASX companies: results land on the ASX platform at market open (10am AEST) or after close (4pm AEST). The ASX announcements page is the primary source.
- For small/mid-cap ASX stocks with 1-3 analysts, a "beat" or "miss" vs. thin consensus is less meaningful than for large-caps. Note this explicitly.
- Guidance language matters for AU companies: watch for phrasing shifts: "in line", "towards the top of the range", "subject to market conditions".
- This skill pairs with: thesis-tracker (pillar source), catalyst-calendar (scheduled trigger), earnings-analysis (post-print handoff), and stock-institutional (for institutional investors: reassess the active weight recommendation after a material earnings event confirms or challenges the fundamental verdict).
Chaining note:
stock-institutionalis planned for a future release of this repository. The earnings workflow in this repo is complete without it.
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.