15. The Cipher-Gears
Domain: Cryptography — Encryption (Symmetric and Asymmetric) POV: Kip Reading time: ~15 minutes
The message arrived on a Tuesday morning, slipped under the door of the shed in an envelope sealed with wax that bore no imprint. Inside was a single sheet of paper, and on the paper were words that Kip could not read.
Not because they were written in a foreign language — the letters were ordinary Citadel script, the same alphabet that Kip had learned when he was five years old and had been reading every day since. But the letters were arranged in an order that made no sense. Words that should have been familiar were jumbled into nonsense. Sentences that should have conveyed meaning conveyed only confusion. It was as though someone had taken a perfectly ordinary message and scrambled it, deliberately and systematically, into something that looked like language but was not.
Kip stared at the paper for a long time, turning it over in his hands, holding it up to the light, searching for some pattern or clue that would explain what he was seeing. Nothing. The message was gibberish — intentional, careful, expertly constructed gibberish.
“It is encrypted,” Quill said, when Kip brought the message to the emergency council. “The words have been transformed using a cipher — a mathematical operation that scrambles the original text into something unreadable. Only someone who possesses the correct key can reverse the transformation and recover the original message.”
“Who sent it?” Sable asked.
“I do not know. There is no signature, no seal, no identifying mark of any kind. But the fact that it was delivered to the shed — a place that only the five of us and Penric know about — suggests that whoever sent it knows who we are and where to find us.”
“Vale,” Wren said immediately.
“Perhaps,” Quill said. “Or perhaps someone who is trying to warn us about Vale. The encryption itself is a clue — whoever wrote this message was trying to ensure that only the intended recipient could read it. They were protecting its confidentiality. The question is: who is the intended recipient, and what is the key?”
The Library’s restricted stacks contained several treatises on the art of cipher-making, and Quill — who had finally been granted Level Three authorization after the council’s reforms — spent the next day retrieving and studying them. What he found was a tradition of cryptography that stretched back to the Citadel’s founding, a tradition built on two fundamental approaches.
The first was symmetric encryption: a single key, shared between the sender and the recipient, that could both scramble and unscramble the message. The sender used the key to transform the plaintext into ciphertext; the recipient used the same key to reverse the transformation. It was fast, efficient, and as old as the city itself. But it had a fatal weakness: the key had to be shared, and the sharing created a vulnerability. If the key was intercepted during transmission, or if either party was compromised, the entire system collapsed.
The second was asymmetric encryption: a pair of keys, mathematically linked but not identical. One key — the public key — could be shared freely with anyone. The other key — the private key — had to be kept absolutely secret. A message encrypted with the public key could only be decrypted with the private key. A message encrypted with the private key could only be decrypted with the public key. The system was more complex than symmetric encryption, and slower, but it solved the key-sharing problem: the private key never needed to leave its owner’s possession.
“The message we received,” Quill said, spreading the treatises across the workbench in the shed, “could have been encrypted with either method. If it used symmetric encryption, we need the key — a single key, shared by the sender and whoever they intended to receive the message. If it used asymmetric encryption, we need the private key — the one half of a key-pair that was never meant to be shared.”
“How do we find out which one?” Kip asked.
Thread, who had been studying the message in silence for the past hour, spoke for the first time. “Look at the structure of the ciphertext. Symmetric ciphers usually produce output that is roughly the same length as the input — one scrambled word for each original word. Asymmetric ciphers often produce output that is longer, because the mathematical operations involved require padding and additional structure. This message is slightly longer than a typical message of its apparent length would be. I would guess asymmetric.”
“So we need the private key,” Sable said. “A key that was never shared. A key that only one person possesses.”
“Or possessed,” Thread said quietly. “Whoever sent this message may not be alive to share the key. They may have sent it as a last act — a final communication, encrypted so that only the right person could read it, in case the message was intercepted.”
The private key, it turned out, was hidden in the one place that only the fellowship would think to look.
Thread found it three days later, embedded in the silver wire of the Restricted Archive’s door — not as a physical object but as a pattern, a specific configuration of the wire’s shifting geometries that, when interpreted correctly, produced a sequence of symbols that matched the structure of an asymmetric decryption key. The door had been holding the key all along, waiting for someone who knew how to recognize it.
“The door was configured to ask three questions,” Thread explained. “Identity, authorization, context. But it was also configured to hold a secret — a key that could decrypt messages encrypted with a specific public key. The sender of our message must have known about this. They must have known that the door would be the only place where the key could be safely stored and the only place where we would think to look.”
“Who would know that?” Kip asked.
Thread was silent for a moment, and then: “The person who configured the door. The person who taught it to ask better questions. The person who left the new book on the reading table, with the inscription ‘Ask better questions’ on its first page.”
“The door itself,” Quill breathed. “The door has been helping us. Not just by asking better questions — by holding the key to a message that was meant for us.”
They retrieved the key-pattern from the door’s silver wire — a delicate process that Thread performed with the same patient precision that Thread brought to everything — and Quill applied it to the encrypted message, following the procedures described in the Library’s cryptographic treatises. The transformation was slow — asymmetric decryption was always slower than symmetric — but when it was complete, the original message appeared on the page as clearly as if it had been written that morning.
It was a single sentence, written in a hand that none of them recognized:
Vale has a master key.
The implication took a moment to sink in, and when it did, the silence in the shed was absolute.
A master key. A single cryptographic token that could unlock every door, decrypt every message, override every access control in the city. If Vale possessed such a key — if he had somehow obtained or created a credential that bypassed all the new security measures they had implemented — then everything they had built over the past weeks was compromised. The split keys, the three questions, the verification checkpoints, the zero-trust architecture — all of it could be circumvented by someone who held a key that the system was designed to trust absolutely.
“We need to change the keys,” Kip said, and his voice was steadier than he expected. “All of them. Every lock, every cipher, every access token in the city. If Vale has a master key, we need to make sure that key no longer works.”
“That will take weeks,” Sable said. “Months, perhaps. The city’s cryptographic infrastructure has been built up over centuries. Every door, every archive, every secure communication channel depends on keys that were generated and distributed and trusted over generations. Changing all of them — and making sure that the new keys are secure — is not something we can do overnight.”
“Then we start now,” Wren said. “And we do not stop until it is done. Because if Vale can open any door he wants, whenever he wants, then none of our other defenses matter. The walls, the barriers, the checkpoints — they are all bypassable. The only thing standing between Vale and the heart of the city is a set of keys that he already has.”
The key rotation — the systematic replacement of every cryptographic credential in the Citadel — became the fellowship’s primary project for the next two weeks. It was, as Sable had predicted, an immense undertaking: thousands of locks needed new keys, thousands of archives needed new encryption, thousands of access tokens needed to be invalidated and replaced. The work was tedious and unglamorous and absolutely necessary, and it taught Kip something he had not expected to learn: that cryptography was not just about mathematics and algorithms and the clever scrambling of letters. It was about trust — about the assumption that a key in the right hands was a guarantee of security, and that a key in the wrong hands was a guarantee of disaster.
“Encryption protects confidentiality,” Quill said, as the fellowship worked through the endless lists of keys that needed to be rotated. “It ensures that even if a message is intercepted, the interceptor cannot read it without the key. But that protection is only as strong as the key management. If the keys are not protected — if they are shared carelessly, stored insecurely, or allowed to fall into the wrong hands — then the encryption is worthless. The strongest cipher in the world cannot protect a message whose key has been compromised.”
“So the work we are doing now,” Kip said, “is not just about replacing keys. It is about making sure that the new keys are managed better than the old ones were. That no single person ever holds a master key again. That every key is protected — stored securely, shared only when necessary, rotated regularly — so that even if one key is compromised, the damage is limited to what that key can unlock, and nothing more.”
Quill looked at him with something that might have been surprise, or respect, or both. “Yes. That is exactly what we are doing. Key management — the generation, storage, rotation, and destruction of cryptographic keys — is the foundation on which all other cryptographic protections rest. And it is the most neglected part of the entire system, because it is the least visible and the least glamorous and the easiest to ignore until something goes wrong.”
Kip looked at the list of keys in front of him — doors and archives and communication channels, thousands of them, each one representing a point of vulnerability that needed to be closed — and he understood, with a clarity that had been growing since the moment he pressed his ear to the cold patch on the Wall, that the work they were doing was not just about securing the city against Vale. It was about building a system that could not be compromised by any single person, no matter how trusted or how clever — a system where trust was distributed and verified and rotated, where no one held the keys to everything, and where the compromise of any single key did not mean the compromise of the whole.
The message had been encrypted to protect its confidentiality. But the real protection was not in the cipher — it was in the key, and in the system that managed the key, and in the people who understood that even the best cryptography in the world was only as strong as the trustworthiness of the people who held the keys.