Dennis Ross, a distinguished fellow at the Washington Institute, served in senior national security positions in the Reagan, George H.W. Bush, Clinton and Obama administrations.

Israeli intelligence scored another coup this week. The Israeli release of thousands of Iranian documents proving that Iran was actively developing a nuclear explosive device is now a source of intense debate. Unfortunately, the debate falls along predictable lines: Supporters of the Iran nuclear deal say the revelations prove the importance of the deal in stopping Tehran’s march to a bomb. Opponents of the deal claim the opposite: The Iranian intention to possess nuclear weapons remains unchanged, and because the deal permits the Iranians to have a large nuclear infrastructure, it will enable them to produce a bomb later, when the restrictions in the agreement lapse.

Ironically, both sides have a point, but neither is listening to the other. They are too caught up in trying to influence President Trump’s decision to stay in the Joint Comprehensive Plan of Action (JCPOA) — as the agreement is officially known — or leave it on May 12. The proponents of the agreement should not be so quick to simply dismiss all the documentation.

Yes, maybe we knew the Iranians had tried to develop a weapon, testing neutron initiators that are critical for producing a chain reaction, but we did not know how extensive the preparations were. Nor did we know that the Iranians had saved all the files on their research and development of a bomb and moved them to a secret location well after the JCPOA entered into force in 2016. What does that say about the Iranian commitment in the deal never to seek, acquire or develop a nuclear weapon? If that commitment was real, why would such materials not be destroyed? It is hard to escape the conclusion that the Iranians were preserving the option of developing a bomb later.

That raises the question of whether the Iranians were violating the JCPOA. Section T of the deal commits the Iranians not to “engage in the following activities which could contribute to the development of a nuclear explosive device: Designing, developing, acquiring, or using computer models to simulate nuclear explosive devices.” Does moving files that simulate such devices constitute “using” computer models? It certainly is preserving them.

Supporters of the JCPOA need to address these questions and not dismiss them, arguing that only the deal can stop Iran from resuming its nuclear program of enrichment and weaponization. Indeed, supporters must also address the issue that the vaunted transparency provided by the nuclear deal is more limited than we need it to be. If nothing else, the scope of the inspections needs to be wider to provide greater confidence that the Iranians have not resumed their weapons-related activities.

But the opponents of the deal cannot just rail against it and say we must walk away from it. What then? Maybe the Israeli disclosures and the materials will convince the British, French and Germans that the deal requires corrections. Though less likely, the Russians and Chinese, too, are likely to have been surprised by the scope of the Iranian weaponization efforts and could be open to certain adjustments. Still, at this point, none is inclined to give up on the deal and scrap it.

So the opponents of the deal should offer a strategy on what to do. Logically, the revelations give the parties to the agreement a reason to go back to the Iranians and insist on an explanation of these materials. They should demand verifiable destruction of all these files — lest they give the Iranians the ability to pick up quickly where they left off on weaponizing. And, given the danger that Iran would have the ability to move rapidly to a weapon in the year 2030, when the limits on enrichment, centrifuges and their output, and reprocessing end, the sunset issue needs to be addressed.

True, the sunset provisions represented the core trade-off in the deal: reductions in the number of Iranian centrifuges and constraints on the Iranian nuclear enrichment program in return for an end to those limits by 2030. But it is clearly also a fact that the Iranians have been preserving the means they had developed for building a nuclear bomb — and that betrays the purpose of the deal. It is not enough to ensure that they have stopped their program and the stoppage can be verified. It is also essential, at a minimum, to buy more time for the limitations on Iranian enrichment. Even if the Iranians are not willing to accept these limitations in perpetuity, the revelations may well create leverage with the Europeans to press for a 10-year extension of the sunset provisions until 2040 — and Iran, in these circumstances and with signs of genuine unrest at home, might not be so ready to say no to the Europeans.

The Trump administration should use the Israeli revelations as a new source of leverage to “fix” the deal and not simply walk away from it. At this point, if the president walks away on May 12, he is still likely to walk away alone. Our pressure on Iran has been most effective historically when we have acted with our allies and not on our own. The Israeli intelligence coup offers a new opportunity to do so.