1. Future commodity prices.  It is reasonable to use the 3-yr historical average.  Now if you are bullish on a commodity then obviously you are going to use your prediction of future prices in trying to value a mining asset.  However, it is a bad sign if the report’s author is using some number that is higher than the 3-yr historical average.  The author is trying to stretch the numbers to make the mining asset look more attractive.  If the author is using commodity prices provided by the mining company’s management… run away.  And if the commodity price is supposed to go up due to inflation… run away.
2. Exchange rates.  If it’s not the exchange rate at the time the report was written… run away.  This is a bogus method for inflating commodity prices or deflating costs.
3. NPV at a discount rate of 5% or lower.  A discount rate like that is stretching.  If the resource is clearly economic, the author would probably start with a base case of 7.5-10%.
4. Geological interpretation.  The size of a deposit could be exaggerated by using an extremely large search size.  Deposit size can also be inflated by interpolating between drill intersections that are incredibly far apart.
5. Untested metallurgical process.  What works on a bench-scale may not necessarily work in a full-scale production environment.  Granted, this isn’t necessarily a red flag as it can make sense to try newer and better recovery techniques.  However, there are technological risks involved that may lead to cost overruns.
6. Were previous technical reports bogus?  Go on SEDAR.ca and look at the technical reports for previous deposits.  In the case of Canada Lithium, they released a bogus report for one of their gold properties claiming several hundred thousand ounces of (Inferred) resources.  Then they raised capital.  Then this property was written down to almost 0 in the same quarter.

Unfortunately, there are also subtle ways to manipulate reserves and NPV that are very difficult to spot.

1. Aggressive engineering assumptions.  A devious author could feign ignorance to engineering complications and leave costs out.  For example, KWG thought that a railroad to its deposit would cost $900M.  Now it is saying that it would cost almost $2B.
2. Grade interpolation method.  This can make a difference of 20% depending on whether nearest neighbour, ID², ID³, etc. are used.  It is a good sign if the report shows the results for many interpolation methods.
3. Mine engineering inherently involves a number of moving parts.  Subtle adjustments to the various factors can have a huge impact on mine economics.  For example, a 5% adjustment to deposit size, mining recovery and metal recovery can lead to a ~15.7% increase in revenue.  A 5% change to operating+capital costs for a mine with a 80% margin would lead to a 20% increase in profit.  It is actually slightly higher than that if lower-grade ore becomes economic due to the lower operating costs.  The bottom line is that a series of very small changes can have a cumulative effect on mine economics.