The Temple of the Architect was less a temple and more a fortress that had forgotten it was supposed to be welcoming.
Massive stone walls, narrow windows, defensive architecture that suggested the goddess of creation had strong opinions about security. The interior matched—stark corridors, minimal decoration, everything built with structural integrity as the primary concern over aesthetics.
Perfect place to hide from people trying to kill you.
The priesthood had taken one look at our bedraggled group of fugitives at three in the morning and invoked sanctuary rights without asking questions. One of the benefits of worshipping the deity who'd literally built the world—her priests had authority over physical spaces, and sanctuary laws were absolute.
Even the City Watch couldn't touch us here.
Probably.
"The Architect's sanctuary has held for three thousand years," the head priestess—an older woman named Mother Cassandra—had told us while showing us to the temple's guest quarters. "No authority, mortal or divine, may violate it without her express permission. You are safe here."
"What about Executors?" I'd asked.
She'd paused. "If Executors come, we will remind them of the law."
Not exactly reassuring, but better than nothing.
That had been three days ago.
Three days of lying low. Three days of waiting for the inevitable knock on the door that would mean someone had found us. Three days of every footstep in the corridor making my heart rate spike.
But the knock hadn't come.
So I'd done what I always did when stressed and trapped somewhere: I'd studied the system.
---
The temple library was extensive. Apparently when your goddess was the one who'd designed the world's underlying architecture, her followers accumulated a lot of technical documentation.
Spell theory texts. Magical engineering treatises. Even something that was basically a style guide for proper spellcasting technique, which read like the world's most pompous coding standards document.
I'd read all of it.
And now I was going through the Gray Zone's personal spell collection, methodically decompiling every single one.
"You've been at that for six hours," Corvina said, appearing in the doorway of the small study room the temple had given us. "Have you eaten?"
"Pip brought me something earlier." I gestured vaguely at a plate that might have had food on it at some point. "I'm cataloguing."
"Cataloguing what?"
"Everything." I turned the page of the spellbook in front of me—handwritten notes from some previous Gray Zone member, probably decades old. "Every spell structure I can find. Building a mental repository. Cross-referencing common patterns, identifying frequently used functions, mapping out the standard library of magical operations."
Corvina pulled up a chair, sat down across from me. "Why?"
"Because understanding the common patterns makes it easier to spot the uncommon ones. If I know how most spells are structured, I can identify the interesting ones faster—the exploits, the backdoors, the poorly implemented functions that have unintended side effects." I activated Code Vision on the current spell. "Like this one. Basic fire blast. Watch."
The spell structure materialized in my vision:
```
FUNCTION: cast_fire_blast()
ALLOCATE mana_cost = 25
CREATE projectile
element = FIRE
damage = 3d6
velocity = 15 m/s
lifetime = 5 seconds
collision_detection = TRUE
LAUNCH projectile(target_direction)
IF collision_detection THEN
APPLY fire_damage(target)
CREATE fire_effect(target, duration=3)
ENDIF
END FUNCTION
```
"Looks standard to me," Corvina said.
"It is. Which is the problem. Look at the lifetime parameter—five seconds. This spell allocates resources for a projectile that will exist for five seconds, but the projectile moves at fifteen meters per second and combat ranges are rarely more than thirty meters. Which means the projectile reaches its target in two seconds, but the spell is paying for five seconds of existence."
"So?"
"So that's wasted mana. The spell should calculate optimal lifetime based on distance to target, not use a hardcoded value." I pulled out a piece of paper—I'd been taking notes in actual ink, since paper didn't leave digital traces that could be tracked. "And that's just one inefficiency. Most combat spells have three to five optimization opportunities. Defensive spells have even more. The entire magical ecosystem is running on bloated legacy code."
Corvina was quiet for a moment. Then: "Can you write better ones?"
"Better versions of existing spells? Absolutely. New spells from scratch? Maybe. I'd need to understand the underlying system better—the actual runtime environment, not just the user-space functions." I closed the spellbook. "But yes. Given time and mana to experiment, I could write spells that are thirty to fifty percent more efficient than standard implementations."
"That's..." She trailed off, looking for the word.
"Broken," I supplied. "I'm broken. Or the system is. One of those."
"I was going to say 'valuable.'" Corvina leaned back. "You realize mages spend years—decades—learning optimization techniques. And you've been here three days and already surpassed most of them."
"I'm not doing anything special. I'm just reading the source code. The information was always there—you all just couldn't see it."
"Because we're not NULL class anomalies who break reality as a hobby."
I smiled despite myself. "As a compulsion. There's a difference."
"Is there?"
"Hobbies are optional."
She laughed—actual laughter, not the nervous kind from earlier in the week. It was progress. The Gray Zone had been tense for days, waiting for the other shoe to drop, waiting for Executors to storm the temple, waiting for everything to go wrong.
But nothing had gone wrong yet.
Which meant we had time to prepare for when it inevitably did.
"Thorne wants to talk to you," Corvina said, standing up. "About training. Real training, not just reading books."
"Combat training?"
"He thinks you need to learn how to defend yourself without relying entirely on magic. Says your exploitation abilities won't help if someone stabs you before you can deploy them."
I considered arguing that I'd been doing fine with pure technical exploitation so far, but Thorne had a point. I'd nearly died multiple times in the last week. A little practical combat knowledge might improve my survival probability.
"Fine," I said. "Where is he?"
"Courtyard. Try not to get hurt."
---
The temple courtyard was designed for meditation and contemplation. Stone pathways between carefully maintained gardens, a central fountain, benches placed for optimal aesthetic appreciation of the architecture.
Thorne had turned it into a training ground.
He'd cleared a section of paving stones, marked out a practice area with chalk, and was currently stretching in a way that suggested someone who'd done this routine a thousand times before.
[NPC: THORNE CORVUS]
Level: 12
Class: SHADOWBLADE [UNCOMMON]
Status: READY FOR TRAINING
Current Mana: 78/95 MP
Active Buffs: ENHANCED_FLEXIBILITY [23 minutes]
Threat: NONE [FRIENDLY]
"You're late," he said without looking up.
"I was cataloguing spell structures."
"Of course you were." He straightened, tossed me a wooden practice sword. "Ever used one of these?"
I caught it awkwardly. The weapon felt wrong in my hands—too heavy, too long, center of balance all off. "Does 'no' count as an answer?"
"It's the answer I expected." He drew his own practice sword in one smooth motion, settled into a ready stance. "Basic lesson: combat is pattern recognition. Attack patterns, defense patterns, timing patterns. You're good at patterns."
"I'm good at patterns in code. I don't know if that translates to—"
He attacked.
Not hard. Not trying to hurt me. But fast enough that I barely registered the movement before the wooden blade was coming at my shoulder.
I tried to block. Failed. The practice sword tapped my shoulder with the kind of gentle impact that said "you just died."
Thorne reset to his starting position.
"Again," he said.
This time I was watching with Code Vision active. Saw his combat AI execute the attack pattern:
```
ATTACK_SEQUENCE: overhead_strike_basic
Step 1: WEIGHT_TRANSFER(forward)
Step 2: ARM_EXTENSION(45_degrees)
Step 3: BLADE_PATH(overhead_arc)
Step 4: TARGET_ZONE(shoulder)
Execution_time: 0.8 seconds
```
I saw it coming this time. Knew exactly where the blade would go.
Still couldn't block it.
My body knew what to do but couldn't execute fast enough. Knowing the pattern and having the muscle memory to counter it were completely different things.
The practice sword tapped my other shoulder.
"You're seeing it," Thorne observed. "I can tell. You know what I'm about to do. But you can't move fast enough."
"Pattern recognition doesn't help if I don't have the physical coordination to respond."
"Exactly. Which is why we're training." He reset again. "Your Code Vision is an advantage, but advantages only matter if you have the fundamentals to exploit them. Right now you're like someone who can see all the possible chess moves but doesn't know how the pieces move."
Made sense. Frustrating, but made sense.
"So we drill fundamentals," he continued. "Over and over. Until your body learns the patterns as deeply as your brain sees them. Then your Code Vision becomes a force multiplier instead of just interesting trivia."
We drilled.
For two hours straight, Thorne ran me through basic sword movements. How to hold the weapon. How to stand. How to move your feet without losing balance. How to block high, block low, block left, block right.
Basic. Repetitive. Boring.
But I could see the patterns emerging in my own movements. Could see my combat AI starting to form—crude, inefficient, full of bugs, but there. My body learning to execute simple responses to simple threats.
[SKILL GAINED: SWORD COMBAT - BASIC]
[Allows: Simple attack/defense patterns]
[Success rate: LOW but improving]
"Better," Thorne said after I successfully blocked three attacks in a row. "You're learning."
"My arms feel like they're filing a harassment complaint."
"That means they're actually doing work for once." He lowered his practice sword. "Take five. Drink water. We'll run drills again in a few minutes."
I collapsed onto a bench, breathing hard, every muscle in my upper body staging a protest. But my Code Vision was still active, still analyzing.
And I was watching Thorne's movements even at rest.
The way he shifted his weight. The micro-adjustments in stance. The constant environmental awareness. His combat AI was always running, always processing threat information, always ready.
Muscle memory encoded as living code.
"Can I ask you something?" I said.
Support the author by searching for the original publication of this novel.
"Sure."
"Why are you helping me? All of you. The Gray Zone. You're risking everything—sanctuary violation, execution, divine attention—to protect someone you met less than a week ago."
Thorne was quiet for a moment. Sat down on the bench across from me.
"Because you're right," he said finally. "About the system. About the corruption. About all of it." He looked down at his hands. "My father was a noble's bastard. Not acknowledged, not legitimate, but enough noble blood that I grew up seeing how the system works from both sides. Saw how different the rules were depending on which side of the legitimacy line you fell on."
"That's why you became a Shadowblade."
"That's why I became someone who specializes in operating outside the rules." He met my eyes. "The system is broken, Hex. We've all known it for years. But we didn't have proof. Didn't have someone who could see the actual code and point to the specific bugs. You're not just a criminal we're hiding. You're evidence that the system can be understood. Analyzed. Fixed."
"Or crashed."
"Or that." He smiled without humor. "But either way, you're the first person I've met who looks at magic the same way I look at nobility—as a system built by people, maintained by people, and therefore vulnerable to people who understand how it really works."
Made sense. Thorne had his own grievances with the system. So did Corvina. So did all of them. I wasn't just a fugitive they were protecting—I was a symbol that the system wasn't invulnerable.
Which was a lot of pressure, actually.
"I'm not sure I can fix it," I said quietly. "The system, I mean. The corruption goes really deep. Divine-level deep. I'm Level Three. Even with exploits, there's only so much I can do."
"You're Level Three now," Thorne corrected. "Who knows what level you'll be when this is over. If you keep growing at the rate you've been growing..." He trailed off. "Let's just say I've never seen anyone level up from pure analysis before. Most people need to kill things to gain experience. You gained two levels from reading documentation and corrupting files."
"The system seems to reward me for understanding it rather than fighting it."
"Which is exactly the kind of user the administrators would fear most." He stood up, stretched. "Ready for another round?"
I wasn't. My arms were still protesting. But I stood up anyway.
"One more thing," I said. "When you're attacking—when your combat AI is executing patterns—can you feel it happening? Or does it just feel like instinct?"
Thorne considered the question. "Instinct, mostly. I don't think about individual movements. I just... react. Why?"
"Because I can see it as code. Every decision you make, I can see it queue up before you execute it. And I'm wondering if I can learn to predict it well enough to counter it even without matching muscle memory."
He smiled. "Let's find out."
We drilled for another hour. And this time, I wasn't just learning sword movements. I was learning to read combat as code—to see the attack patterns forming in Thorne's AI, to predict his next three moves based on his current stance and previous actions, to position myself where I needed to be half a second before I needed to be there.
It wasn't elegant. It wasn't efficient. But it worked.
And when I finally managed to land a hit—just once, probably because he let me—Thorne's grin suggested I'd passed some kind of test.
[SKILL IMPROVED: SWORD COMBAT - BASIC → COMPETENT]
[Pattern-based prediction compensating for lack of training]
[Unique combat style developing: ANALYTICAL FIGHTER]
"That's enough for today," Thorne said, genuinely impressed. "You're exhausted, I'm exhausted, and we both need food."
I nodded, too tired to argue.
But as we walked back toward the temple interior, Corvina met us in the corridor with an expression that meant new problems.
"Hex," she said. "We need to talk. About mana."
---
Mana efficiency theory turned out to be significantly more complicated than I'd expected.
Corvina had set up a small experimental area in one of the temple's unused rooms—chalk circles on the floor, monitoring crystals positioned around the perimeter, basic safety measures for magical testing.
"You've been optimizing everyone's spells," she said, arranging the crystals with precise spacing. "Making them more efficient. But you haven't optimized your own mana usage."
"I've barely cast anything. Most of my abilities are analysis-based, not active casting."
"Exactly. Which means you haven't learned your limits." She activated the crystals—they began glowing with soft light, monitoring ambient mana levels. "You have a bigger pool than most mages at your level. One-forty is closer to what a Level Ten would have. But you don't know how to push it."
"Push it?"
"Overclock. Burst casting. Drawing more power than your pool can sustainably provide." She demonstrated, casting a simple light spell but pouring excessive mana into it. The light blazed painfully bright for three seconds, then cut off. "Trade-off: more power for less duration. Useful in emergencies."
I thought about it. Made sense from a systems perspective—trading sustained performance for burst capacity, accepting reduced efficiency for increased peak output.
"But there's a risk," Corvina continued. "Push too hard and you burn out. Literally. Your mana channels can't handle the throughput. You pass out, or worse."
"Define worse."
"Permanent damage to your magical capacity. I've seen mages who overclocked too often—they never recovered their full pools. Stuck at fifty, sixty percent of what they should have."
That sounded like hardware failure from overclocking a processor. Increased voltage, increased heat, reduced lifespan. Same principles applied to magic, apparently.
"Show me how to do it safely," I said.
"There is no 'safely.' That's the point. You learn your limits by approaching them carefully and hoping you recognize the warning signs before you cross them."
"So we experiment with increasing mana loads until I find the failure threshold?"
"Basically yes. Start small. Cast something simple but push extra mana into it. Feel what happens. Stop if anything feels wrong."
I moved into the center of the chalk circle. Activated Code Vision to monitor my own processes—I could see my mana pool, the regeneration rate, the current allocation.
```
MANA_POOL: 140/140 MP
REGEN_RATE: 1 MP / 30 seconds
SAFE_CASTING_THRESHOLD: ~50 MP per spell
CURRENT_LOAD: 0%
WARNING_SYSTEMS: ACTIVE
```
Cast a basic light spell. Normal cost: 15 MP.
But I pushed more into it. 20 MP. Then 25. Then 30.
The light grew brighter with each increase. And I could feel the difference—not pain, but pressure. Like running a program that was consuming more resources than recommended, the system heating up, fans spinning faster.
At 35 MP, the pressure became noticeable discomfort.
At 40 MP, my vision started blurring at the edges.
At 45 MP, the warning systems activated:
```
WARNING: MANA_THROUGHPUT EXCEEDING SAFE LIMITS
CURRENT_DRAW: 45 MP (300% over baseline)
CHANNEL_STRESS: MODERATE
RECOMMENDED: REDUCE LOAD IMMEDIATELY
```
I cut the spell. The light died. My mana pool showed 95/140, and I felt... not burned out, but definitely strained. Like I'd just sprinted up stairs.
"How did that feel?" Corvina asked, watching the monitoring crystals.
"Like my internal processes were complaining about resource allocation." I waited for my vision to clear. "But I could push further if I needed to. The warning systems kicked in at forty-five, which means the actual failure threshold is probably higher."
"Don't test it today. Overclocking damage is cumulative—you might not notice it until you've done permanent harm." She deactivated the monitoring crystals. "Practice the safe threshold first. Learn to consistently push to thirty or forty percent overclock without hurting yourself. Master that before you try anything more aggressive."
Made sense. Gradual stress testing rather than immediate maximum load.
We spent the next hour practicing. I cast different spells—light, force, basic elemental effects—and practiced overclocking each one by consistent amounts. Building muscle memory not just for casting, but for pushing past safe limits in controlled ways.
[ABILITY UNLOCKED: MANA OVERCLOCK]
[Push mana beyond safe limits for burst performance]
[Increased power at cost of channel stress]
[Risk of burnout if overused]
[Current safe limit: 40% overclock]
By the time we finished, I was exhausted again, my mana pool sitting at 68/140, and Corvina was looking at me with that expression that meant she was impressed but trying not to show it.
"You learn too fast," she said. "It's unsettling."
"I'm just reading the system documentation."
"You're exploiting the system in ways that should require years of training." She started gathering the monitoring crystals. "Has anyone ever told you that you're basically a security researcher in a world that doesn't know security research exists?"
"That's the most accurate description anyone's given me since I got here."
She smiled. "Get some rest. We'll train more tomorrow."
---
That night, I couldn't sleep.
Too much information. Too many patterns. Too many problems with no obvious solutions.
So I sat in the temple library, surrounded by spellbooks and technical manuals, and I did what I always did when stressed: I analyzed systems.
The Covenant Operating System. The divine administrators. The noble privilege structure. The magical architecture underlying everything.
It was all connected. All part of a single complex system that had been running for ten thousand years without proper maintenance.
And the more I understood it, the more obvious the fundamental problem became:
The system wasn't just corrupt. It was failing.
The memory leaks Thorne had mentioned. The inefficient spell implementations. The way nobles needed to buy system access to maintain stability during "crisis periods."
These weren't just symptoms of corruption. These were symptoms of a system that was literally falling apart, and the administrators were patching it with duct tape and bribes because they didn't understand the underlying code well enough to actually fix it.
The gods weren't malicious. They were incompetent.
They'd inherited a system they didn't fully understand, they'd been maintaining it through trial and error, and now it was breaking down faster than they could patch it.
And they were terrified someone would notice.
That's why NULL class users were so dangerous. Not because we were powerful—because we could see the bugs. Could point to the specific places where the system was failing. Could prove the gods weren't omnipotent administrators but confused inheritors running someone else's code.
I pulled out my notes. Started mapping out what I knew:
- Covenant Operating System installed 10,000 years ago by "The Precursors"
- Current pantheon found existing system, claimed credit
- Original architects uploaded as system core (imprisoned AIs?)
- System failing due to lack of understanding of source code
- Noble privileges are bandaids on critical errors
- Memory leaks accumulating
- Magic becoming less efficient over time
- Divine administrators panicking
The solution was obvious: Get access to the original source code. Understand how the system was built. Fix the actual problems instead of applying temporary patches.
But that meant getting into the Root Directory. The palace archives. Divine-level restricted access.
Which meant I needed to level up. Significantly.
I was Level Three. I needed to be... what? Level fifty? Sixty? High enough that I could actually survive divine attention long enough to do something useful.
That would take time. Experience. Real progression.
The door opened behind me. Pip, carrying two cups of tea, looking like he hadn't slept either.
"Thought you might be here," he said, offering me a cup.
I took it. "Can't sleep?"
"Thinking about everything." He sat down across from me, looked at my notes. "You think we can actually fix this? The whole system?"
"I think someone needs to try. And I seem to have been volunteered by being the only person who can see the bugs."
"That's not fair."
"No," I agreed. "But it's accurate."
We sat in comfortable silence for a while, drinking tea, both thinking about impossible problems.
"Hex?" Pip said quietly. "When you fix things—when you hack into systems and change them—are you scared?"
"Terrified. Every time."
"But you do it anyway."
"Fear is just my risk assessment system working properly. Doesn't mean I should ignore the vulnerabilities just because they're dangerous to exploit."
He thought about that. Nodded.
"I'm glad you're here," he said.
"Even though I brought divine attention down on all of you?"
"Especially because of that." He smiled—genuine, optimistic, the kind of smile that suggested he hadn't yet learned to be properly cynical. "You're the first person who's made me think we might actually win."
I wasn't sure we would win. The odds weren't good. The opposition was literally divine. The system was failing and we were fighting against people who had power we couldn't match.
But we had something they didn't: We could see the bugs. Could understand the failures. Could exploit the weaknesses.
Maybe that was enough.
"Get some sleep," I told Pip. "We've got a long road ahead."
He nodded, stood up, left me alone with my notes and my tea and my impossible task.
Outside, the temple bells rang midnight.
Three days of sanctuary. Three days of training.
How much longer until the Executors found us?
And when they did, would I be strong enough to survive?
I looked at my character sheet one more time:
```
ALEXANDRIA "HEX" VOLKOV
Level: 3
Mana: 140/140 MP
Abilities: Code Vision, Decompile, Patch, Injection, Overclock
Skills: Hacking (Expert), Combat (Competent), Analysis (Master)
Status: Wanted by gods
Threat: Maximum
```
Not strong enough yet.
But getting there.
One exploit at a time.
---
**EXPERIENCE GAINED: COMBAT TRAINING +200 XP**
**EXPERIENCE GAINED: MANA THEORY +250 XP**
**EXPERIENCE GAINED: SYSTEM ANALYSIS +300 XP**
**LEVEL UP!**
**ALEXANDRIA "HEX" VOLKOV is now LEVEL 4**
**Stat Increases:**
- Mana: +20 (140 → 160 MP)
- Processing Speed: +15
- Code Vision Range: +2 meters (11 meters)
**Skills Improved:**
- Sword Combat: Basic → Competent
- Pattern Recognition: Master → Expert Master
- Mana Control: None → Intermediate
**New Ability Available:**
Choose one:
- [ENHANCED DECOMPILE] - Faster analysis, deeper insights
- [BOTNET BASICS] - Create first zombie process from defeated enemy
- [CRYPTOMANCY FOUNDATIONS] - Begin learning defensive encryption
I selected ENHANCED DECOMPILE. More information was always better than more power.
**Ability Upgraded: DECOMPILE → ENHANCED DECOMPILE**
- Analysis speed increased by 50%
- Can now see optimization opportunities automatically
- Reveals hidden parameters in spell structures
- Success rate improved against complex magic
Three days down. Unknown days remaining until Executors arrived.
And I was four levels into understanding a system that had been running for ten thousand years.
This was going to take a while.
---
**STATUS UPDATE — END OF CHAPTER 7**
**ALEXANDRIA "HEX" VOLKOV**
- Level: 4 [+1]
- Class: NULL [UNDEFINED BEHAVIOR ENABLED]
- Location: TEMPLE OF THE ARCHITECT - SANCTUARY
- Status: TRAINING, SAFE (FOR NOW)
**Mana: 160/160 MP [+20]**
**XP: 900 / 5,000**
**Trace Risk: 67% [EXECUTORS SEARCHING BUT SANCTUARY HOLDING]**
**New Abilities:**
- MANA OVERCLOCK [ACTIVE] — Push beyond safe limits for burst power
- Current safe threshold: 40% overclock
- Risk of burnout if overused
- Cumulative damage possible
- ENHANCED DECOMPILE [ACTIVE - 20 MP] — Upgraded from basic version
- 50% faster analysis
- Shows optimization opportunities automatically
- Reveals hidden parameters
- Better success against complex magic
**Skills Improved:**
- Sword Combat: Basic → Competent [Pattern-based fighting style]
- Mana Control: None → Intermediate [Can safely overclock]
- System Analysis: +1 [Understanding system failures]
**Character Development:**
- Integrating with Gray Zone as team member
- Learning physical combat to complement technical skills
- Understanding scope of system failure (not just corruption)
- Accepting responsibility for fixing broken world
**Training Progress:**
- 3 days in sanctuary
- Catalogued 50+ spell structures
- Developed basic combat competence
- Learned safe overclocking techniques
- Identified root cause: System is failing, not just corrupt
**Key Insight:**
Gods aren't malicious—they're incompetent administrators running inherited code they don't understand. System installed 10,000 years ago by Precursors, now breaking down due to lack of proper maintenance.
**Current Objective:**
- Continue training and leveling
- Wait for Executor threat to pass (or arrive)
- Prepare for eventual palace infiltration
- Fix broken system (somehow)
**SYSTEM NOTE: User is methodically building competencies.**
**SYSTEM NOTE: Training montage successful.**
**SYSTEM NOTE: But Executors are still coming.**
**SYSTEM NOTE: Tick tock.**
