- Home
- Skills
- Research & Analysis
- Elite Eye
Works with the AI Agents you already use
Elite Eye
Evaluate factual, scientific, and marketing claims using evidence grading and calibrated Bayesian probability.
Free
Elite Eye
Example session with this skill installed
Audit the claim that taking 1000mg of Vitamin C daily prevents the common cold in healthy adults. Give me a probability and evidence scoreboard.
- Read your context and instructions
- Compiled the elite eye
The claim has been evaluated against clinical trial data and systematic reviews regarding prophylactic use in general populations.
Reality Check: Vitamin C and Cold Prevention
Claim tested: Daily 1000mg Vitamin C supplementation prevents the onset of the common cold in healthy, non-stressed adults.
- Short verdict: Mostly unsupported.
- Confidence: High (based on large-scale meta-analyses).
Reality vs. Marketing: While Vitamin C is essential for immune function, the "prevention" narrative significantly overstates the effect for the general population.
Evidence Scoreboard
| Evidence | What it actually shows | Quality | Limitations |
|---|---|---|---|
| Cochrane Meta-analysis (29 trials) | No significant reduction in incidence in general population; small reduction in duration (8%). | A | Includes varied dosages; high heterogeneity in some subgroups. |
| Specialized Population Studies | 50% reduction in cold risk for marathon runners and soldiers in sub-arctic conditions. | B | Not generalizable to sedentary or average-activity adults. |
| Pharmacokinetic Data | Plasma saturation occurs at ~200-400mg; excess 1000mg doses are largely excreted. | A | Individual absorption rates vary slightly. |
Incentive and Funding Audit
Most historical support stems from early, non-randomized observations (e.g., Pauling). Current industry funding for supplement trials often focuses on secondary markers rather than clinical incidence. No evidence of systemic fraud, but significant "publication bias" towards small, positive duration studies.
Probability / Bayesian Scoreboard
- Target event: 1000mg/day prevents a cold that would otherwise occur in a healthy adult.
- Estimate: 5-10% (consistent with a small reduction in duration, not prevention).
Sensitivity: If the subject is under extreme physical stress (e.g., ultra-marathoner), the probability of benefit increases to ~50%.
Next steps
- Review the Cochrane Library summary for specific duration effects in children vs. adults.
- Check if your physical activity levels qualify as "extreme stress" for higher efficacy.
- Evaluate cost-effectiveness of 1000mg doses versus dietary intake of 200mg.
Connects securely to your tools. The creator never sees your data.
About this skill
Falsehoods and exaggerations thrive in the gap between a marketing claim and the underlying data. This skill provides a rigorous, evidence-first framework for auditing factual, scientific, health, and financial claims by separating what is known from what is merely marketed.
What it does
- Claim deconstruction breaks down complex narratives into neutral, independently testable propositions.
- Incentive auditing identifies financial, political, or reputational conflicts of interest without inferring corruption from motives alone.
- Evidence grading categorizes sources from A (Strong) to D (Weak) based on methodology, replication, and data transparency.
- Bayesian synthesis applies calibrated probability updates to claims when defensible numeric inputs are available.
- Reality vs. Hype analysis maps the gap between public slogans and the actual evidence-supported utility.
How it works
- Deconstruct the user's input to identify the literal claim, implied causal links, and omitted caveats.
- Audit the empirical evidence by searching for systematic reviews, trials, or primary documents while checking for funding biases.
- Synthesize a verdict using a multi-role analytical pass that includes adversarial review to search for counterevidence.
- Report the findings using a structured scoreboard that includes a factuality index, evidence quality grades, and sensitivity analysis.
Frameworks & tools
This skill utilizes Bayesian probability models (prior odds, likelihood ratios), evidence-based medicine (EBM) grading standards, and formal logic for identifying cognitive biases and logical fallacies.
Why this beats prompting it yourself
Standard prompts often return a binary "true or false" or mirror the user's existing skepticism. This skill enforces a multi-pass analytical protocol that explicitly searches for null results, models source dependence to avoid double-counting, and prevents the fabrication of certainty where evidence is missing.
Use cases
- Fact-checking a viral health trend or nutritional supplement claim against clinical data.
- Evaluating a corporate policy or economic prediction by auditing the funding behind the supporting whitepapers.
- Auditing a product's marketing slogans to determine the actual magnitude of real-world benefit.
- Estimating the likelihood of a geopolitical or technological event using calibrated probability ranges.
Known limitations
This skill cannot provide personalized medical, legal, or financial advice. Accuracy is strictly limited to available data; if browsing tools are unavailable, it cannot verify real-time events.
How to use
Skills on Agensi run inside your AI agent. Copy the prompt, paste it into your agent, and it connects to Agensi and installs this skill for you. If a payment is needed, your agent shows it to you first.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill passes our trust checks below.
Security scanned
Passed our 8-point scan before listing
Fresh listing
Recently published to Agensi
Free forever
No account required to browse
Trust & safety
Security scanned
Verified clean today
- Free to download with an account
Requires