- Home
- Skills
- Trading & Investing
- Stock Research & Indicator Audit
Stock Research & Indicator Audit
It separates reported results, guidance, news and interpretation.
Free
Works with the AI Agents you already use
Stock Research & Indicator Audit
Example session with this skill installed
Research the fictional DEMO Manufacturing stock using the supplied company-facts.md and stock-input.json. Calculate daily indicators, explain the exact screening conditions and identify missing valuation inputs. Do not browse or treat the fixture as a real stock.
- Read your context and instructions
- Compiled the stock research
DEMO Manufacturing — offline synthetic research
All company figures and prices below are invented test inputs. This is not a real issuer, market report or backtest.
Identity: SYNTHETIC:DEMO, VND/share and volume in shares. Research as of 2026-01-15; daily input covers 260 explicitly completed synthetic weekdays. These dates do not represent an exchange holiday calendar.
Company and financial context
The supplied fictional business manufactures industrial components. Both comparison periods are consolidated full fiscal years. Arithmetic was executed from the operands in examples/company-facts.md:
| Metric | Operands | Result |
|---|---|---|
| Revenue YoY | FY2025 120bn / FY2024 100bn - 1 | +20% |
| Attributable profit YoY | 12bn / 10bn - 1 | +20% |
| Attributable profit margin | 12bn / 120bn | 10% |
| FY2025 ROE | 12bn / mean(50bn,70bn) | 20% |
| Simple FCF proxy | CFO18bn - cash capex8bn | 10bn VND |
| P/E | TTM diluted EPS missing | N/A |
| Market capitalization | Current shares outstanding missing | N/A |
Source: examples/company-facts.md, synthetic publication 2026-01-15. No external news or issuer disclosures were supplied; no current news claims can be established.
Executed technical helper — full output
Daily stock indicator audit
SYNTHETIC:DEMO | VND | VND/share | volume shares
As of 2026-01-15T10:00:00Z; timezone Asia/Ho_Chi_Minh; basis unadjusted.
Source assertion: Synthetic fixture, not market prices; weekdays do not represent an exchange calendar.
Bars: 260 closed of 260; excluded 0; 2025-01-17 to 2026-01-15; age 0 calendar days.
| Indicator | Value |
|---|---|
| close | 359 |
| change1Pct | 0.27933 |
| sma20 | 349.5 |
| sma50 | 334.5 |
| sma200 | 259.5 |
| rsi14 | 100 |
| macdLine | 7 |
| macdSignal | 7 |
| macdHistogram | 0 |
| macdPreviousHistogram | 0 |
| atr14 | 3 |
| atr14Pct | 0.835655 |
| bb20Middle | 349.5 |
| bb20Upper | 361.032563 |
| bb20Lower | 337.967437 |
| prior20High | 360 |
| prior20Low | 338 |
| prior20VolumeMean | 1000 |
| volumeRatio20 | 2 |
| return20Pct | 5.899705 |
| Explicit condition | Result |
|---|---|
| closeAboveSma20 | true |
| sma50AboveSma200 | true |
| rsiAbove70 | true |
| rsiBelow30 | false |
| macdBullishCross | false |
| macdBearishCross | false |
| closeAbovePrior20High | false |
| volumeAbove1_5xPrior20 | true |
Daily closed bars; SMA-seeded EMA; Wilder RSI/ATR; RSI flat=50; ATR first TR=high-low; Bollinger population std (ddof=0); prior20 excludes current; comparisons use unrounded values.
- Unadjusted OHLC: review corporate-action discontinuities before interpretation.
- Source, closed flags, session completeness and adjustment consistency are input assertions, not independently verified.
- Conditions describe observations; thresholds are illustrative, not validated trading rules.
Interpretation and watch conditions
The close is above SMA20 and SMA50 is above SMA200, which describes the supplied upward price series. RSI14=100 reflects that every synthetic close increased; it does not establish an imminent reversal. MACD line and signal are both7, histogram0: there is no bullish crossover. Volume is2x its prior20-session average, but close359 remains below the prior20-session high360, so the rolling-high breakout condition is false.
No composite score or buy/sell instruction is justified. For real research, the next inputs are an identified issuer/exchange, verified price units and adjustment policy, a licensed consistent daily history, the actual comparable filings and dated material news. The helper output and arithmetic above were executed; live-data accuracy and financial facts were not independently established because this is a synthetic fixture.
Connects securely to your tools. The creator never sees your data.
About this skill
Company facts and technical numbers you can check
Ask your agent to research a stock and get a brief with security identity, dated company disclosures, financial context, material news, and technical conditions backed by reproducible arithmetic.
Research workflow
The skill guides your agent to verify ticker and exchange, distinguish VND from thousands of VND, track fiscal periods and report units, and attach sources to material claims. It separates reported results, guidance, news and interpretation. P/E stays unavailable when a suitable EPS input is missing or nonpositive; partial-year numbers are not silently treated as full-year results.
Included calculation helper
The dependency-free Node.js helper validates one daily OHLCV JSON file, excludes incomplete bars and calculates SMA20/50/200, Wilder RSI14 and ATR14, SMA-seeded MACD12/26/9, Bollinger20/2, prior-20-session volume ratio and rolling extrema. It shows explicit true/false/unknown conditions. A MACD state is not mislabeled as a crossover, and current-session volume is excluded from the comparison average.
Why this is useful
A convincing stock report can still contain invented prices, a mixed-unit valuation or an indicator calculated on too little history. This skill makes those assumptions visible and keeps missing values unknown. Indicator conventions and source references are bundled, with a report template and a complete synthetic example.
Access and limits
Free. No market-data subscription, brokerage connection or API key is included. Your agent uses available public sources or documents/data you supply. The helper does not fetch quotes or parse financial filings. Current research needs browsing/data access; unavailable history means technical values stay pending.
Automated calculations support daily JSON only. Intraday data, automatic split adjustment, holiday-calendar verification, portfolio advice and order execution are outside this release. Screening thresholds are illustrative observations, not a validated strategy or return guarantee. The helper passed 12 tests on Node 24.15.0 / Windows; other agent clients have not been independently tested.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Trading & Investing.
Daily JSON input only for automated calculations, up to 10 MiB and 10,000 rows. No bundled data feed, automatic financial-statement extraction, split adjustment or exchange-calendar validation. Input provenance/closed flags must be verified. Missing history yields unknown values. Vendor outputs can differ with seeds and source adjustments. No investment-performance guarantee.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Free to use
No account required to browse
Trust & safety
Security scan
Security scan passed 22 days ago
- Free with an Agensi account
- Software
- Browserorweb search tool
- Optional
Node.js 18+
The browser or search tool finds current public-company research. Node.js runs the optional local indicator helper.
Permissions
Uses available public company/market sources or user-supplied documents and daily OHLCV. Browser/network access is for public research; the bundled helper itself makes no network calls. Local read access plus temporary helper/input/report writes may be needed. No brokerage connection, secrets or environment variables are required.