We report on a multiwavelength observation of the blazar 3C 454.3 (which we dubbed crazy diamond) carried out on November 2007 by means of the astrophysical satellites AGILE, INTEGRAL, Swift, the WEBT Consortium, and the optical-NIR telescope REM. Thanks to the wide field of view (FoV) of the AGILE satellite and its prompt alert dissemination to other observatories, we obtained a long (three weeks), almost continuous γ-ray coverage of the blazar 3C 454.3 across fourteen decades of energy. This broad-band monitoring allows us to study in great detail light curves, correlations, time-lags and spectral energy distributions

We report on the second Astrorivelatore Gamma a Immagini Leggero (AGILE) multiwavelength campaign of the blazar 3C 454.3 during the first half of 2007 December. This campaign involved AGILE, Spitzer, Swift, Suzaku, the Whole Earth Blazar Telescope (WEBT) consortium, the Rapid Eye Mount (REM), and the Multicolor Imaging Telescopes for Survey and Monstrous Explosions (MITSuME) telescopes, offering a broadband coverage that allowed for a simultaneous sampling of the synchrotron and inverse Compton (IC) emissions. The two-week AGILE monitoring was accompanied by radio to optical monitoring by WEBT and REM, and by sparse observations in midinfrared and soft/hard X-ray energy bands performed by means of Target of Opportunity observations by Spitzer, Swift, and Suzaku, respectively. The source was detected with an average flux of ∼250 × 10 −8 photons cmabove 100 MeV, typical of its flaring states. The simultaneous optical and γ -ray monitoring allowed us to study the time lag associated with the variability in the two energy bands, resulting in a possible one-day delay of the γ -ray emission with respect to the optical one. From the simultaneous optical and γ -ray fast flare detected on December 12, we can constrain the delay between the γ -ray and optical emissions within 12 hr. Moreover, we obtain three spectral energy distributions (SEDs) with simultaneous data for 2007 December 5, 13, and 15, characterized by the widest multifrequency coverage. We found that a model with an external Compton on seed photons by a standard disk and reprocessed by the broad-line regions does not describe in a satisfactory way the SEDs of 2007 December 5, 13, and 15. An additional contribution, possibly from the hot corona with T = 10 6 K surrounding the jet, is required to account simultaneously for the softness of the synchrotron and the hardness of the IC emissions during those epochs.