equity-research
Skill0 starsChanged 4 months agoApache-2.0View on GitHub
Produces a standalone, user-independent assessment of whether a single equity is attractive on its own merits — DCF intrinsic value, quality ratios, and trading multiples from free public data (SEC EDGAR + market quotes). Use when the user asks whether a specific stock is worth buying or selling. Read-only; never executes a trade and never gives investment advice.
7 other projects use a version of this.
Adopted by
Most recently taken up by
- vijaysuperpositionist/skillstheir own version, first seen Nov 1, 2026
- shadowandy/equity-researchtheir own version, first seen Oct 25, 2026
- The-Utopia-Studio/skillstheir own version, first seen Oct 13, 2026
- anthropics/financial-services-pluginstheir own version, first seen Oct 8, 2026
- Lirunyi-123/equity-research-skilltheir own version, first seen Sep 2, 2026
- rollingSirius/equity-research-skilltheir own version, first seen Aug 12, 2026
Other versions7 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 7
- anthropics/financial-services★ 41ksame name
- anthropics/financial-services-plugins★ 41ksame name
- rollingSirius/equity-research-skill★ 429same name
- The-Utopia-Studio/skills★ 6same name
- Lirunyi-123/equity-research-skill★ 1same name
- shadowandy/equity-research★ 0same name
- vijaysuperpositionist/skills★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository8 other files it gives its agents.
- AGENTS.mdAGENTS.md
- action-draftingSkill
- goals-onboardingSkill
- portfolio-analysisSkill
- researchSkill
- reviewerSkill
- spending-analysisSkill
- suitabilitySkill
The file
Files
equity-research
Purpose
Answers the security-only half of "should I buy/sell this stock?" — is the
name attractive on its own merits, independent of any user. It produces an
EquityAssessment:
a DCF-derived intrinsic value and upside/downside versus the current price,
fundamental quality ratios, the subject's trading multiples, a standalone
valuation verdict (undervalued / fairly_valued / overvalued / unknown),
and factual drivers/risks.
Its output feeds the suitability skill, which pairs this security-only view
with the user's IPS, holdings, and risk tolerance to produce an advisory
recommendation. This skill has no notion of the user.
Methodology adapted from the Apache-2.0 anthropics/financial-services
reference skills (model-builder/dcf-model, model-builder/comps-analysis) —
see the repository NOTICE.
When this skill runs
- The user asks whether a specific ticker is worth buying or selling (e.g. "is AAPL a good buy right now?"), or asks to analyze/value a name.
- The root planner selects it before
suitabilitywhen a recommendation is requested.
It does not run for imperative trade commands with an explicit quantity
("buy 10 shares of AAPL") — those go to action-drafting.
Deterministic core + advisory narrative
Per the reviewer pattern (ADR-0014), the numbers of record are deterministic:
primitives/equity_research.py computes the DCF, ratios, and multiples from a
FundamentalsSnapshot. The LLM layer only narrates and contextualizes those
numbers — it never invents figures. Any divergence is auditable.
Inputs
| Field | Source | Required |
FundamentalsSnapshot | executors.EdgarFundamentalsProvider (SEC EDGAR, primary/free), behind the TTL cache | Yes |
| Latest price | executors.get_quote (Alpaca free market data; deterministic mock fallback offline) | Recommended — needed to compute upside and a verdict |
ValuationAssumptions | Defaults in the primitive; overridable (discount rate, terminal growth, projection years) | No |
Data access scope: read:external_market_data, read:fundamentals. This skill
reads only public company data — no Firestore/BigQuery, no user data — so its
tool surface can never leak private financial information (same isolation
rationale as research).
Output: EquityAssessment
Transient (not persisted), like the Drift Report and Research Brief. Key fields:
| Field | Meaning |
dcf.intrinsic_value_per_share_usd, dcf.upside_pct | Two-stage DCF value and its gap to the current price |
quality | Net/FCF margin, revenue CAGR, ROE, debt/equity from the latest annual period |
multiples | Subject's P/E, P/FCF, EV/EBIT, market cap |
valuation_verdict | undervalued / fairly_valued / overvalued / unknown |
confidence | high / medium / low by data completeness |
key_drivers, key_risks | Factual, model-derived points |
disclaimers | Always present — this is not investment advice |
Failure modes
- No fundamentals / unknown ticker → the provider raises; the skill reports that it cannot assess the name rather than fabricating figures.
- Missing FCF or price → the DCF is skipped and the verdict is
unknown; quality/multiples are still returned where computable. Guessing is never substituted for missing data.
Registry metadata
- Registered as:
projects/{project}/locations/{location}/skills/private-equity-research - Skill revision: 0.1.0 (draft — not yet registered)
- Approval scope:
read:external_market_data,read:fundamentals
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.