- Home
- Skills
- Trading & Investing
- MarketClarity Pro — Stock Research & Risk Desk
More screenshots
Works with the AI Agents you already use
MarketClarity Pro — Stock Research & Risk Desk
MarketClarity Pro A serious stock-research workflow for agents that need to show their work.
$75
MarketClarity Pro — Stock Research & Risk Desk
Example session with this skill installed
Use MarketClarity Pro to audit the bundled ILLUSTRATIVE / NOT LIVE candidate file and shadow-observation file. Calculate all four scores using the published weights, list the execution vetoes, and show which names qualify or are rejected. Then compare the synthetic signal-to-close moves against timestamp-matched VOO and VTI marks. Explain the missed-winner categories. Do not invent real prices, recommend trades, or claim these observations are realized returns.
- Read your context and instructions
- Compiled the marketclarity pro
MarketClarity | Synthetic Session Audit
Evidence notice: ILLUSTRATIVE / NOT LIVE. MC-A, MC-B, MC-C, and MC-D are fictional demonstration symbols; all prices and factor ratings are invented test inputs. These are not stock picks, actual market results or executed paper trades.
The helper checked the four weighted score formulas and market gate values against the supplied records. It did not independently authenticate a price vendor.
Symbol Core Investment Active Growth Asymmetry Swing Trade Research disposition
MC-A 96.0 96.0 96.0 96.0 Qualified research setup; numerical gates pass
MC-B 96.0 96.0 96.0 96.0 PASS — hard veto
MC-C 88.0 88.0 88.0 88.0 Watchlist; below 90 swing floor
Rejection trace: MC-A's synthetic quote, spread, volume, stop and 3:1 reward:risk meet the encoded conditions, subject to authenticating the underlying inputs. MC-B scores equally highly but remains rejected because the source record flags a major binary event during the expected holding period. MC-C has no explicit veto but its 88.0 swing score fails the 90.0 eligibility threshold. No order is placed and no rejected name receives a trade recommendation.
The missed-mover review — observation only
Symbol Original research status Signal→close VOO same window VTI same window Interpretation
MC-A Qualified +3.00% +0.33% +0.30% On qualified research list
MC-B Discovered—Rejected +7.00% +0.33% +0.30% A later rise does not erase the original event risk
MC-D Not Discovered +10.00% +0.33% +0.30% Investigate upstream candidate discovery
Key insight: The +10% synthetic MC-D observation is a discovery question, not a losing trade by the qualified shortlist. The +7% MC-B observation is a veto tradeoff, not permission to rewrite the rules after seeing a price chart. VOO/VTI comparisons use the same synthetic timestamps. None of the above percentages represents a live or simulated filled position.
Next research step: Accumulate many real, timestamp-aligned observations including losers, missing data and costs. Evaluate missed-mover recall and out-of-sample performance before proposing any scoring changes. No historical outperformance has been demonstrated.
Connects securely to your tools. The creator never sees your data.
About this skill
MarketClarity Pro A serious stock-research workflow for agents that need to show their work.
Most stock prompts end with a confident-sounding list of tickers. MarketClarity Pro asks the harder questions: What evidence supports each score? How fresh is the quote? Does the setup actually pass the risk gate? And what did the shortlist miss?
MarketClarity Pro equips a compatible AI agent with a repeatable research process for screening, comparing, challenging, and auditing stock candidates. It combines four independent, published scoring models with strict vetoes and timestamp-based evidence checks. It is designed to make abstaining, rejecting, and learning from misses as deliberate as selecting a candidate. What makes it different
- Four genuinely distinct lenses. Core Investment (long-term ownership), Active Growth (business momentum and valuation), Asymmetric Opportunity (upside versus permanent-loss risk), and Swing Trade (entry, invalidation and execution). Each has explicit deterministic factor weights. No artificial “best stock = 95” normalization.
- Hard risk vetoes. A mathematical score does not bypass poor liquidity, wide spreads, missing evidence, unresolved binary events, weak reward:risk or lack of a valid stop. A 96/100 can still be rejected.
- Evidence-first research. Quote timestamps, source attribution, spread, volume, market session and missing fields are visible. Unverifiable results become N/A, not invented certainty.
- A separate missed-mover investigation. Track what was never discovered, what was discovered but rejected, and what qualified. Analyze those outcomes without quietly loosening a disciplined scoring system.
- VOO and VTI as permanent context. Compare observations to both funds on matching timestamps, while keeping hypothetical price marks separate from actual or paper trading returns.
- Concrete reusable tools. Includes the scoring specification, evidence contract, session-by-session research workflows, reporting templates, synthetic fixtures, a standard-library Python audit script and automated tests.
- Readable explanations. Each unfamiliar indicator is explained with its definition, measurement, relevance, interpretation and common misconception. Example tasks
- Run an early premarket discovery report, then narrow to a small qualified research watchlist.
- Audit five candidates and explain why the highest number did not pass.
- Review stocks that surged after the open and identify whether discovery or qualification missed them.
- Build a transparent source-timestamp and VOO/VTI benchmark note.
- Determine whether available evidence is sufficient to discuss a hypothetical paper-trade setup. What is included
- SKILL.md: complete agent instructions and hard constraints
- references/scoring-model.md: factor weights, bands, vetoes and conservative operational checks
- references/data-contract.md: required input schema, timestamps, source quality and missing-data rules
- references/research-workflows.md: early premarket, primary ranking, final pre-open, regular-session and post-close procedures
- references/report-templates.md: buyer-facing research, rejected-setup and missed-mover layouts
- scripts/marketclarity_audit.py: deterministic arithmetic and gate checks; no external libraries
- scripts/test_marketclarity_audit.py: eight automated checks
- examples/: original synthetic candidate and shadow-ledger fixtures Requirements and boundaries The bundled examples run offline with Python 3.9+ if your agent can execute local scripts. The methodological instructions can also be used by agents without Python. For current stock research, supply an authorized, timestamped market-data source and relevant company information; no real-time feed or data license is included. A host agent may need browser/network access to retrieve sources.
This is an agent research skill, not the hosted MarketClarity app, an investment advisory service, automated trading software, or a guaranteed signal provider. It does not place live or paper orders. Its scoring weights are an initial transparent framework, not a claim of proven predictive power or outperformance. Output is research and hypothetical scenario analysis, not individualized financial advice.
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.
No live trading or brokerage execution. Requires external data sources. Incomplete data results in N/A or Insufficient Evidence status. Does not provide financial advice.
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
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 12 days ago
Nothing to install. It runs in your AI agent.
- Optional
Python 3.9+Timestamped market data
Python runs the offline audit helper. Your own market data is only needed for current-market research.