Can you share which of the space telescope proposals are you talking about? Several of them are 4-8x the mass of JWST, and up to 2-4x times the length of JWST's minimum payload faring even folded up.

Space assembly is going to be required even for a number of the proposals components i mentioned above, or at least multiple heavy lift launches in the star-shield senario.

In engineering space launch systems/vehicles, it isn't just 100 tons everywhere in the solar system, its 100 tons to a specific apogee/orbital altitude/delta V target. That is why Falcon heavy can only get 17.637 tons to Mars (assuming the least fuel consumptive Holman transfer window), but can get 69.4 tons to LEO. https://www.spacex.com/vehicles/falcon-heavy/ This is why Falcon Heavy missed bids on a number of deep space missions as it had to reserve fuel mass to hit the orbital targets and could only hit those numbers in non-re-useable mode (Another reason Ariane 6 was designed to hit SpaceX's disposable price per kg numbers for the Falcon Heavy).

Worth mentioning, re-entry velocity from high LEO/GEO/TLI/L2 is very different compared to re-entry from LEO. Typical low earth orbit re-entry speeds are near 17,500 mph and the Mach number M 24-25 for the lower end of LEO orbits. For Artemis/Apollo/L2 return in days/weeks, it is 24,500 mph (Mach 32). This is another reason why L2/TLI/Helio is a different ballgame when it comes to human rated craft payload and mission price per kg. The asteroid sample return mission this year is i think going to be a record breaker, but coming from another solar orbit entirely would be the high end for re-entry velocity/heat abatement. Stardust sample return re-entry from a deep space location with 28,856 mph (Mach 38, 12.9 km/sec) holds the record for the fastest reentry of any human made object, but would be the upper end of a human based refit mission.

Starship is aiming to get 100 tons of payload (Payload includes mission Life support, deep space shielding, mission/fuel/food/water/air/etc), in 1,000 cubic meters of payload volume to a optimal 98.9° orbital inclination(less optimal inclinations means much less payload to LEO), with an altitude of 310 mi (500km), traveling at a minimum of 11 orbits a day of velocity. Space X has said it will not invest in hypersonic testing facilities, so i want to add that may mean more payload lost as starship will need to move to tiles the same thickness as the space shuttle's as the Hubble servicing missions were at the extreme limit of the LEO orbital band and speed. [It used to be 150 tons but it was reduced as the real performance of the raptors became known, and SpaceX completely scrapped active cooling in exchange for buying the shuttle TPS system from NASA about 2 years ago.] https://upload.wikimedia.org/wikipedia/commons/b/b4/Comparison_satellite_navigation_orbits.svg