본문으로 건너뛰기
cccskills
무료GitHub에 공개

stock-analysis

Research public companies and listed securities with current market, company, ownership, and filing data through SandBase. Use for ticker analysis, price-performance comparisons, insider activity, or an evidence-backed investment research brief; do not use for trade execution or personalized financial advice.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

설치 방법 보기

포함된 파일(2)

  • SKILL.md6.2 KB
  • references/sandbase-api-map.md3.0 KB

SKILL.md(원문)

설치하기 전에 에이전트에게 전달되는 지시 내용을 확인할 수 있습니다.

Stock Analysis

Build a current, traceable research view of a public company or listed security. Separate sourced facts from calculations and interpretation, state the data timestamp, and avoid presenting a directional opinion as personalized financial advice.

Read the SandBase API map before retrieving live data. It defines the supported MCP and REST transports and the rules for dynamic capability discovery.

Scope the request

Resolve the following from the request or state reasonable defaults:

  • security name, ticker, exchange, and market region;
  • question to answer: company overview, price behavior, valuation context, ownership, filings, comparison, or full brief;
  • analysis window and comparison benchmark;
  • currency and whether prices should be adjusted for splits and distributions;
  • output depth and whether the user wants raw data, a concise answer, or a reusable report.

Disambiguate tickers before making paid calls when the symbol could refer to multiple exchanges or instruments. Do not silently treat an ADR, index, fund, option, or cryptocurrency as common stock.

Retrieve evidence through SandBase

Choose one available transport from the API map. Use SandBase for live external data; do not call the original Manus runtime or substitute a provider-specific credential.

  1. Discover capabilities for the specific evidence needed, such as company profile, price history, analyst or valuation context, holders or insider transactions, and regulatory filings.
  2. Inspect every selected capability immediately before execution. The live schema and pricing are authoritative; model names and parameters must not be guessed from examples or prior runs.
  3. Use the smallest capability set that answers the question. Before a multi-call paid batch, summarize the selected endpoints, call count, and known or estimated cost for confirmation.
  4. Execute only the approved calls. Preserve each response, source timestamp, endpoint name, and run ID in an evidence ledger.
  5. Poll an asynchronous run by its returned ID until it reaches a terminal state. Never create a duplicate call merely because a run is still pending.

If neither SandBase transport is available, do not fabricate current prices or company facts. Ask for a user-provided export, or provide an analysis template that clearly marks live-data fields as unavailable.

Select the minimum dataset

RequestEvidence to retrieve
Company overviewIdentity, exchange, industry, business description, reporting currency, and latest profile timestamp
Price or trendTimestamped OHLC/adjusted-close series, volume when relevant, corporate actions, and chosen benchmark
Technical contextPrice series sufficient for the requested indicators; compute indicators transparently when raw data is available
Valuation or analyst viewCurrent metric definitions, reporting period, peer or sector context, rating date, and source provenance
Ownership or insider activityHolder or transaction records, role, transaction type, date, quantity/value, and reporting basis
FilingsForm type, filing date, reporting period, issuer identity, and source document URL
ComparisonThe same fields, currency treatment, interval, and date window for every security

Do not require an unrelated profile or analyst call when one focused dataset is enough. For broad due diligence, gather at least company identity, price history, the latest relevant filing, and the risk factors needed to support the requested conclusion.

Normalize and calculate

  • Preserve raw values and units. Record currency, timezone, exchange session, and whether prices are adjusted.
  • Align comparison series to common trading dates and a common starting value before calculating relative performance.
  • Calculate returns from the selected price convention and show the formula or method. Do not mix adjusted and unadjusted series.
  • Treat missing values, stale quotes, suspended trading, corporate actions, and differing fiscal calendars explicitly.
  • Label provider-produced scores, targets, and ratings as third-party estimates with their observation date.
  • Prefer primary filing documents for accounting claims. Use profile or market-data summaries as navigation aids, not as substitutes for the filing when wording matters.

For computed indicators, state the lookback, sampling interval, and whether the current incomplete bar was excluded. Never infer support, resistance, fair value, or insider sentiment from a single opaque score without explaining the evidence.

Analyze without overclaiming

Organize the answer around the user's decision rather than dumping every field:

  1. Identity and as-of time — instrument, exchange, currency, market status, and latest observation.
  2. What changed — price, volume, operating, filing, or ownership signals over the requested window.
  3. Why it may matter — evidence-backed interpretation with plausible alternative explanations.
  4. Risks and gaps — stale or missing data, conflicting sources, concentration, liquidity, event, accounting, or comparability risks.
  5. Next checks — the few facts that would most change the conclusion.

When asked whether a security is a "buy," translate the request into a non-personalized scenario analysis. Present bull, base, and bear considerations or explicit decision criteria; do not claim certainty, guarantee returns, or execute a trade.

Final handoff

Return:

  • a direct answer with the as-of date and market timezone;
  • a compact table of the most decision-relevant facts and calculations;
  • source or filing links when returned by SandBase;
  • the SandBase endpoint names and run IDs used;
  • assumptions, missing data, and material limitations;
  • a clear distinction between sourced facts, derived calculations, and interpretation.

Do not claim a live-data analysis succeeded if any required run is pending, failed, or returned an empty result.

리뷰

아직 리뷰가 없습니다. 사용해 본 소감을 남겨 주세요.

댓글

로그인 없이 쓸 수 있습니다. 사용 후기, 질문, 작동하지 않은 환경 등을 편하게 남겨 주세요.

아직 댓글이 없습니다. 첫 댓글을 남겨 보세요.

보내면 바로 공개됩니다. 이메일 주소나 전화번호 같은 개인 정보는 쓰지 마세요. URL은 2개까지입니다.

0 / 1,000자

이 양식은 reCAPTCHA로 보호되며 Google 개인정보처리방침 및 서비스 약관이 적용됩니다.

같은 저장소의 스킬

Search academic papers, scholarly articles, and research publications through SandBase. Use when asked for literature review, academic citations, scholarly research, paper discovery, or scientific evidence gathering.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

sandbaseai/sandbase-skills2022026년 9월 26일 업데이트

Research emerging academic trends by combining scholarly databases, web sources, and news to identify rising research areas, breakthrough papers, and shifting scientific consensus. Ideal for R&D teams tracking frontier science.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

sandbaseai/sandbase-skills2022026년 9월 26일 업데이트

Create, implement, or review accessibility review work with format, accessibility, performance, and visual-verification controls. Use when the requested deliverable depends on accessibility review.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

sandbaseai/sandbase-skills2022026년 9월 26일 업데이트

Plan, implement, or review agentic eval work in an existing codebase with compatibility, security, and verification controls. Use when the user explicitly requests agentic eval work.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

sandbaseai/sandbase-skills2022026년 9월 26일 업데이트

ai-seo

무료

Create or review ai seo deliverables with audience, evidence, voice, and conversion-quality controls. Use when the user explicitly asks for ai seo work.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

sandbaseai/sandbase-skills2022026년 9월 26일 업데이트

Apply airflow dag patterns in reproducible local data workflows with version-aware APIs, explicit assumptions, and validation. Use when the user chooses airflow dag patterns or its strengths fit the task.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

sandbaseai/sandbase-skills2022026년 9월 26일 업데이트

sandbaseai의 스킬 모두 보기

이 스킬의 문제 신고