Read an agent analysis.
Understand what the agents measure, and what a forecast does and does not say.
The research workflow
A run begins with an asset and a question about its price at a particular horizon. FETCHN gathers available evidence, assigns relevant evidence to specialist agents, and asks the agents for their own outlooks. The pipeline combines valid internal outlooks into a consensus and preserves the individual views, reasons and risks for inspection. Multi-horizon results require at least two valid internal agent outlooks for each reported horizon.
Some providers are optional or can be unavailable during a run. External research can support a specialist’s judgment, but does not become an unreviewed consensus vote. A missing source is a coverage gap, not a positive signal. Open Data sources and verification to see where the evidence can come from and how to inspect it.
What the specialists examine
The standard research team contains seven specialist roles. Each receives evidence relevant to its scope; its output can differ from another specialist’s view of the same asset.
| Specialist | Question it helps answer |
|---|---|
| News | What has recently been reported, and is the source and timing relevant? |
| Social | What are public posts suggesting, and how noisy or one-sided is that signal? |
| How are public community discussions framing the asset? | |
| Research | What do available research materials and company context support? |
| Macro | Could rates, the economy or sector conditions change the outlook? |
| On-chain | What relevant on-chain activity or token data is available? |
| Market | What do price behavior and market conditions say about the proposed direction? |
These are roles in the analysis, not seven guaranteed live data feeds. Coverage depends on the asset, enabled providers and their responses. Read the actual agent breakdown and source links in each result.
Read a result from top to bottom
- Check the asset, generation time and reference price. A result for the wrong symbol or an old result should not drive an order.
- Choose the horizon. “Higher in 24 hours” asks a different question from “higher in 30 days.”
- Read the consensus probability and confidence together, then expand the agent breakdown. A high consensus can conceal wide disagreement or weak evidence from one specialist.
- Read the explanations, cited sources and the listed risks. Check whether a source is recent and whether it actually supports the claim.
- If you intend to trade, open Terminal and request a new executable quote. The research reference price cannot tell you what the testnet pool will pay.
A 68% probability with 61% confidence for 24H means the model estimates the underlying reference price is more likely to be higher at that horizon, with a separate evidence score. It does not mean a 68% chance of a profitable Stock Token swap. Pool price, spread, fees, slippage and liquidity are separate.
Probability and confidence
Probability is the model’s estimated chance that the underlying asset’s reference price will be higher at the selected horizon. It is not the chance that a testnet trade will make a profit.
Confidence reflects the strength and consistency of evidence behind that estimate. The agent can require a minimum for both values before entering a position. Neither number removes market, liquidity or execution risk.
Forecast windows and growth outlook
Markets can show 1H, 24H, 7D and 30D outlooks. Automated Analysis & Trading compares 24H, 7D and 30D together for each Favorite Token. Its growth outlook score summarizes the three probability and confidence pairs on a 0–100 scale. It is a directional research score, not a calibrated probability of profit.
For this strategy, at least two of these three windows must meet your minimum probability and confidence, and none can show a strongly opposing probability below 45%. A qualifying forecast still has to pass balance, permission, pool and slippage checks.
Reference price versus trade quote
The forecast starts from a live reference price for the underlying asset. A testnet trade uses a separate AMM pool quote for the test token. A higher underlying forecast does not imply an executable profit on testnet. Always inspect the displayed quote and liquidity.
Price history, missing charts and fresh signals
Where available, the chart displays indexed DEX trade history from GeckoTerminal. A test token can have a valid contract and wallet balance without enough indexed swaps to form candles. If the chart says market history is unavailable, retry later or inspect the pool; do not fill the gap with the underlying share chart.
Saved forecasts are snapshots, not a continuously updated promise. Agent entry requires a qualifying analysis less than 30 minutes old. The monitoring worker then checks the executable exit value against the strategy’s take-profit and stop-loss rules independently of the original forecast horizon.