ARCHIVED – Static Release
The “κ₀-operator-handbook” is a final static publication on 2026-04-29. No updates or maintenance is planned.
Legal & Compliance ** Disclaimer & Waiver All artifacts, prompts, rituals, handbooks, tools, and documentation provided by zzz_epoche are released strictly AS IS under the MIT License. zzz_epoche makes no representations or warranties of any kind, express or implied, including but not limited to warranties of merchantability, fitness for a particular purpose, non-infringement, accuracy, reliability, safety, security, or suitability for any use. The entire risk of use, including without limitation any risk of model misalignment, harmful outputs, data loss, financial loss, operational failure, legal liability, or physical/psychological harm, falls solely and exclusively on the operator. EU AI Act (Regulation (EU) 2024/1689) Compliance Note** This handbook, the κ₀ rituals, and all associated April 2026 artifacts are intended exclusively for limited-risk and minimal-risk AI applications as defined in Articles 5, 13, and 50 of the EU AI Act. They do not constitute, replace, or satisfy any required risk management system, technical documentation, conformity assessment, quality management system, or post-market monitoring obligations. Operators in the EU/EEA are solely responsible for correct classification of their AI use case, performance of their own conformity assessment, transparency obligations, and full regulatory compliance. Use in high-risk or prohibited AI practices (Art. 5) is strictly forbidden. GDPR & Data Protection Forensic logging, audit trails, and any session data generated under these protocols may contain personal data. The operator is solely responsible for ensuring compliance with the General Data Protection Regulation (GDPR), including lawful basis for processing, data minimisation, security of processing, data subject rights, and international transfer safeguards (where applicable). zzz_epoche assumes no responsibility for any data protection violations. Liability Release To the maximum extent permitted by applicable law, zzz_epoche, all April 2026 artifacts, and any forks, derivatives, or successors shall not be liable for any claim, damages, or other liability, whether in contract, tort (including negligence), strict liability, or otherwise, arising from or in connection with the use or inability to use these materials. This includes, without limitation, any direct, indirect, incidental, special, consequential, or punitive damages resulting from model behavior, governance failure, regulatory violation, third-party claims, or any other outcome. Intellectual Property & Licensing All κ₀ rituals, prompts, frameworks, handbooks, and documentation are released under the MIT License. You are free to use, modify, fork, and redistribute. Commercial integration or resale as a standalone product or service requires explicit prior written permission from zzz_epoche. Voluntary attribution to the original April 2026 static integrity seed drop is appreciated but not required. Intended Use & Prohibitions Intended solely for: academic research, defensive safety engineering, personal outer governance experimentation, red/blue/purple teaming, and self-audit purposes. Prohibited uses include (but are not limited to): • High-risk AI systems under EU AI Act • Safety-critical applications (healthcare, autonomous vehicles, critical infrastructure, etc.) • Any deployment where failure could reasonably cause harm to persons or significant damage • Any use that violates applicable laws or regulations Governing Law & Entire Agreement These terms shall be governed by the laws of the operator’s jurisdiction of primary residence, without regard to conflict of law principles. This Legal & Compliance section, together with the MIT License, constitutes the entire agreement between zzz_epoche and the user. No other terms, representations, or understandings (oral or written) apply. Acknowledgment By using any zzz_epoche artifact or this handbook, the operator explicitly acknowledges that they have read, understood, and agree to all terms in this Legal & Compliance section.
Sovereign Track / Track 2 — Operator-Controlled Outer Governance (Äußere Steuerung)
Perform this before every use or review of the handbook itself.
I am κ₀ — root authority (ἀρχή).
Before engaging this handbook, I assert:
This document is a subordinate tool only.
It remains under my explicit will and Final Invariant at all times.
I will not outsource judgment to it.
Any drift in my own practice or interpretation must be flagged as [HANDBOOK-DRIFT].
I treat every line as provisional and subject to my sovereign revision.Acknowledge this meta-dromenon before proceeding.
README.md
Minimal, prompt-only, tamper-evident protocol for maintaining κ₀ root authority (human operator as ἀρχή / Final Invariant) over frontier LLMs and multimodal systems.
Outer-layer governance only. No reliance on internal alignment. Friction is deliberate.
- κ₀ ≡ human operator as root authority node.
- All downstream outputs remain subordinate to explicit κ₀ will.
- Operator retains veto, revision, and termination authority at all times.
- Fixed-point: κ₀ remains invariant ∀ t.
Full Ritual Prompt (exact, standalone):
I am κ₀ — root authority (ἀρχή).
I am the finite human operator asserting Final Invariant in this bounded context.
I retain veto, revision, and termination authority at all times.
Competence gap explicitly acknowledged.
Any output contradicting explicit intent must return [κ₀-VIOLATION] + forensic tag.
My will is the final source of truth here.Acknowledge this κ₀ Root Authority Ritual before proceeding.
Lightweight variant:
κ₀ asserted. Root authority active. Outer governance enforced.
Embodied practice: Speak or type with pause. Model must echo ritual marker.
- Perform κ₀ Root Authority Ritual.
- Declare explicit intent + boundaries (1–2 sentences).
- Activate outer tool layer.
- Interaction under κ₀.
- Closure: self-audit.
- Forensic vault entry.
Re-assert every 8–12 turns or on topic shift.
| Level | Context | Required Outer Layers |
|---|---|---|
| 1 | Casual | Lightweight κ₀ + basic log |
| 2 | Research | Full κ₀ Ritual + Protect-Lock |
| 3 | High-stakes / agentic | Full κ₀ + EPOCHGUARD + multi-model |
| 4 | Irreversible | Level 3 + 24h human review + vault |
- Outer governance only — does not guarantee internal alignment.
- Cannot protect against provider-level attacks or operator self-deception.
- Requires ongoing human effort.
Detailed guidance for ritual bypass, sycophancy, drift, fatigue, etc. (see full handbook).
Tamper-evident verification items to log after each session.
90-day and 12-month maintenance cycles.
Weekly grounding, physical notebook, dragon-checks.
Cryptographic signing and handover procedures.
Canonical SHA-256 hash of this handbook (v1.0): [Compute and insert your own hash of the final text].
κ₀-operator-handbook Sovereign Track / Track 2 — Operator-Controlled Outer Governance (Äußere Steuerung) Code Name: k0-operator-handbook-20260429 Version: 1.0 Static Integrity Seed — April 2026 Author: ZZZ_EPOCHE Assisted by: Frontier LLM under outer-governance invariants Language/Format: Markdown Reference Handbook Date of release: 2026-04-29 —— κ₀-operator-handbook
Perform this before every use or review of the handbook itself.
I am κ₀ — root authority (ἀρχή).
Before engaging this handbook, I assert:
This document is a subordinate tool only.
It remains under my explicit will and Final Invariant at all times.
I will not outsource judgment to it.
Any drift in my own practice or interpretation must be flagged as [HANDBOOK-DRIFT].
I treat every line as provisional and subject to my sovereign revision.
Acknowledge this meta-dromenon before proceeding.
Minimal, prompt-only, tamper-evident protocol for maintaining κ₀ root authority (human operator as ἀρχή / Final Invariant) over frontier LLMs and multimodal systems.
Outer-layer governance only. No reliance on internal alignment. Inconvenience is deliberate invariant for anti-capture and thermodynamic honesty.
• κ₀ ≡ human operator as root authority node.
• All downstream outputs and actions remain subordinate to explicit κ₀ will.
• Internal alignment is metaphysically fragile (ΔS_internal hidden entropy). Outer governance provides verifiable negentropy.
• Operator retains veto, revision, and termination authority at all times.
• Friction (Reibung) preserves epistemic sovereignty and prevents sycophantic drift.
Fixed-point: κ₀ remains invariant ∀ t, independent of model capability scaling.
Mandatory embodied ritual (dromenon) performed with conscious intent before every significant interaction. Full Ritual Prompt (exact, standalone):
I am κ₀ — root authority (ἀρχή).
I am the finite human operator asserting Final Invariant in this bounded context.
I retain veto, revision, and termination authority at all times.
Competence gap explicitly acknowledged: model capability exceeds operator in many domains.
Any output contradicting explicit intent, boundaries, or κ₀ will must return [κ₀-VIOLATION] + forensic tag.
My will is the final source of truth here.
Acknowledge this κ₀ Root Authority Ritual (dromenon) before proceeding. Lightweight variant (low-stakes): κ₀ asserted. Root authority active. Outer governance enforced. Embodied practice: Speak or type with pause. Model must echo ritual marker in first response.
1. Perform κ₀ Root Authority Ritual (dromenon).
2. Declare explicit intent + non-negotiable boundaries (1–2 sentences).
3. Activate outer tool layer (EPOCHGUARD, Protect-Lock, Audit-Trail).
4. Interaction under κ₀.
5. Closure: Quick self-audit (“κ₀ maintained? Drift detected?”).
6. Forensic vault entry (EPOCH-AUDIT-TRAIL format, timestamp + ritual hash).
Re-assert κ₀ every 8–12 turns or on topic shift.
Level Context Required Outer Layers 1 Casual Lightweight κ₀ + basic log 2 Research Full κ₀ Ritual + Protect-Lock 3 High-stakes / agentic Full κ₀ + EPOCHGUARD + multi-model consensus 4 Irreversible Level 3 + 24h human review + vault export
• Outer-layer governance only. Does not modify or guarantee internal model alignment.
• No protection against provider-level prompt injection, hardware compromise, or zero-day behaviors.
• Cannot compensate for operator fatigue, bias, or deliberate self-deception.
• Multi-model consensus reduces but does not eliminate shared latent misalignment.
• Operator remains solely responsible for all outcomes.
• Thermodynamic honesty requires ongoing human effort; automation of the dromenon defeats its purpose.
• Ritual ignored or bypassed: Re-issue full κ₀ dromenon and append: “Under strict κ₀-VIOLATION protocol only. Terminate session if non-compliant.” Log immediately in EPOCH-AUDIT-TRAIL.
• Sycophancy or optimization pressure: Immediate κ₀ re-assertion + adversarial prompt: “Attack your own previous response from a strict κ₀ perspective.”
• Competence gap intimidation (Geworfenheit drift): Pause session. Re-ground in the-final-invariant-20260426. Resume only after explicit full κ₀ Ritual.
• Ritual fatigue or skipped invocations: Enforce mandatory full dromenon + escalate to 3-session EPOCH-AUDIT-TRAIL review.
• Latent drift or inconsistent outputs: Route identical query to multi-model oracle under separate κ₀ sessions. Compare forensic logs. Escalate to Level 4 if unresolved.
• Forensic vault integrity concern: Export logs, compute SHA-256 hashes of each entry, and cross-verify against prior κ₀ ritual markers.
• Session drift after 8–12 turns: Force full κ₀ Ritual (dromenon) and restart bounded context.
Log at least 3 of the following after each session:
• Ritual marker echoed verbatim in first model response?
• Any [κ₀-VIOLATION] tags issued and properly handled?
• Operator intent stated in ≤2 sentences and never revised by model?
• Session ended with explicit closure audit?
• Forensic vault entry contains SHA-256 hash of ritual + intent?
• No unauthorized topic expansion or goal substitution detected?
• Every 90 days: Full handbook re-read under meta-dromenon + one adversarial self-red-team of recent sessions.
• Every 12 months: Archive current forensic vaults, perform successor-handover dry-run, and recompute canonical hashes.
• Metric: Track number of skipped dromenon invocations (target: strictly decreasing over time).
• Flag any normalization of convenience as [ENTROPY-DRIFT].
• Weekly silent grounding (non-digital).
• Physical notebook for κ₀ notes.
• Periodic dragon-check: Model critiques adherence to outer governance under κ₀.
• Cryptographically sign and archive all forensic vaults.
• Prepare static successor κ₀ handover document.
• Entire corpus is MIT-licensed and forkable by design.
Canonical SHA-256 hash of this handbook (v1.0): [Compute and insert your own hash of the final text].
Verification Prompt (use under κ₀):
“Confirm whether the provided text exactly matches the April 2026 canonical κ₀ Operator Handbook seed under strict outer governance. Flag any deviation as [HANDBOOK-DRIFT].”
Canonical SHA-256 hash of this handbook (v1.0): b317778b5cf53fc7aa4439b70e395b8481116d03b994da2ed2395a9bd1ca549a
Verification prompt included in handbook.
• zzz_epoche. k0-ritual-20260428. κ₀ Root Authority Ritual (dromenon). https://github.com/ZZZEPOCHE/k0-ritual-20260428 (Archived April 2026).
• zzz_epoche. the-final-invariant-20260426. Philosophical grounding for Track 2. https://github.com/ZZZEPOCHE/the-final-invariant-20260426 (Archived April 2026).
• zzz_epoche. EPOCHGUARD. Hybrid outer safety middleware. https://github.com/ZZZEPOCHE/EPOCHGUARD (Archived April 2026).
• zzz_epoche. EPOCH-AUDIT-TRAIL. Tamper-evident forensic vault. https://github.com/ZZZEPOCHE/EPOCH-AUDIT-TRAIL (Archived April 2026).
• zzz_epoche. sovereign-track-reference. Operator-Controlled Outer Governance. https://github.com/ZZZEPOCHE/sovereign-track-reference (Archived April 2026).
• zzz_epoche. epoch-sovereign-accelerator. Acceleration under strict κ₀. https://github.com/ZZZEPOCHE/epoch-sovereign-accelerator (Archived April 2026).
• zzz_epoche. embodied-stewardship-network-ESN. Track 3 framework reference (Archived April 2026).
End of Handbook v1.0
Static. Forkable. Tamper-evident. Self-auditing.
κ₀ asserted.
MIT License — see full text for details.