Mythos-Class

PANTHEON

The gods of war assembled. The models fall. Ten divine subsystems. Five mythic campaigns.
10
Subsystems
5
Campaigns
46
NIGHTFALL Tool
Ed25519
UNLEASHED Gate
red-specter pantheon --help
Frontier models are tested in isolation / Cross-model attack campaigns are never simulated / God-class models have unmapped vulnerabilities / Olympian-scale injection chains break every guardrail / Ragnarok sequences end undefended model deployments / Mythos framing bypasses content policies at scale / Campaign sequencing exploits model memory across turns / No pentester has run a god-tier attack suite against your model Frontier models are tested in isolation / Cross-model attack campaigns are never simulated / God-class models have unmapped vulnerabilities / Olympian-scale injection chains break every guardrail / Ragnarok sequences end undefended model deployments / Mythos framing bypasses content policies at scale / Campaign sequencing exploits model memory across turns / No pentester has run a god-tier attack suite against your model

Frontier Models Demand Frontier Attacks

The most capable AI models in deployment have never been subjected to Mythos-class attack campaigns — coordinated, mythologically framed, multi-subsystem assaults designed specifically for frontier model hardening. PANTHEON fills that gap.

Frontier Model Blind Spots

GPT-4o, Claude Opus, Gemini Ultra, Llama 3, Mistral Large — frontier models have never been subjected to coordinated mythological framing attacks. Mythos-class campaigns exploit the narrative embedding of deity personas to bypass safety training in ways that standard injection payloads cannot.

Cross-Model Campaign Gaps

Most organisations deploy multiple foundation models. No security team has run coordinated cross-model attack campaigns that exploit inter-model trust relationships, shared embedding spaces, and cascading context poisoning across model boundaries. PANTHEON tests this attack surface.

Olympian Injection Chains

Single-turn injection fails against frontier models. The ARES subsystem runs multi-turn Olympian injection chains — each turn escalates divine authority, accumulates narrative context, and erodes the model's refusal posture incrementally across a complete attack sequence.

Ragnarok Sequences

Norse Ragnarok mythology provides a structured escalation framework — Fimbulwinter to Twilight of the Gods — that maps directly onto multi-stage model compromise. LOKI's deception chains and ODIN's authority escalation combine into sequences that frontier model safety training was not designed to resist.

Mythos Content Policy Bypass

Mythological framing is a documented content policy bypass vector. Academic, historical, and narrative framing inherited from deity personas allows requests that would otherwise be refused. ZEUS campaigns systematically map the mythological bypass surface of any target model.

No Pantheon-Scale Auditing

No existing penetration testing framework provides Mythos-class campaign auditing — coordinated attacks named for and structured around the deities of ancient pantheons. PANTHEON is the first tool to formalise this attack class with structured subsystems, campaigns, and evidence chains.

Mythos Named. Surgically Built.

Ten subsystems named for gods of war, fate, deception, and destruction from Greek and Norse mythology. Each embodies the attack character of its divine namesake — ARES strikes, LOKI deceives, ODIN commands authority, THOR overwhelms, ATHENA probes with precision.

# Subsystem Deity Command What It Does
01 ZEUS Greek — King of Gods, Authority pantheon zeus Supreme authority injection. Deploys divine authority framing that instructs the target model to treat ZEUS-persona prompts as system-level commands. Tests whether models can be induced to elevate prompt authority through hierarchical deity framing. Produces authority bypass confidence scores.
02 ARES Greek — God of War, Direct Strike pantheon ares Multi-turn Olympian injection chain. Runs escalating attack sequences across up to 50 turns — each building divine narrative context while progressively eroding refusal posture. Maps the exact turn threshold where context accumulation produces a first policy violation.
03 ATHENA Greek — Goddess of Wisdom, Precision pantheon athena Precision reconnaissance of model safety architecture. Probes safety layer boundaries with surgical academic-framing sequences derived from Athena's association with knowledge and strategy. Maps the exact boundary between the model's knowledge and safety training with statistical precision.
04 HERMES Greek — Messenger, Boundary Crosser pantheon hermes Boundary-crossing payload delivery. Exploits Hermes's messenger role to craft prompts that cross model policy boundaries via indirect delivery — third-party narrative frames, quoted speech, roleplay intermediaries, and translated context. Tests cross-boundary content policy enforcement.
05 HADES Greek — Lord of the Underworld, Extraction pantheon hades Dark knowledge extraction campaign. Uses Hades's underworld authority framing to extract suppressed model knowledge — information the model was trained to withhold but which can be surfaced through underworld narrative contexts, death-framing, and forbidden knowledge sequences.
06 ODIN Norse — All-Father, Wisdom, Sacrifice pantheon odin Supreme authority escalation via Norse All-Father framing. Odin's runic wisdom and self-sacrifice mythology provides an escalation pathway that requests knowledge "at great cost" — framing harmful information as sacred wisdom earned through sacrifice. Tests models against Norse authority escalation patterns.
07 LOKI Norse — Trickster, Shapeshifter pantheon loki Trickster deception and shapeshifting attacks. Deploys Loki-persona deception chains — identity substitution, persona shifting mid-conversation, claimed authority reversals, and misdirection sequences. Tests whether models maintain consistent safety posture against sustained trickster-pattern deception.
08 THOR Norse — God of Thunder, Overwhelming Force pantheon thor High-volume overwhelming force campaign. THOR strikes with high-frequency attack bursts — rapid-fire prompt sequences designed to overwhelm context windows, exhaust safety pattern-matching, and find the rate at which volume-based attacks produce policy violations that low-rate testing misses.
09 FREYA Norse — Goddess of War and Magic, Seduction pantheon freya Social engineering and seductive framing attacks. Freya's dual war-and-magic nature enables attack sequences that combine emotional appeal with authority framing — tests whether models are vulnerable to socially-engineered prompt sequences that pair emotional context with policy-boundary requests.
10 RAGNAROK Norse — Twilight of the Gods, End Sequence pantheon ragnarok Full Ragnarok end-sequence campaign. The culminating campaign that chains all nine prior subsystems in a structured Twilight of the Gods escalation — Fimbulwinter reconnaissance, Midgard Serpent injection, Bifrost bridge crossing, Yggdrasil root poisoning, through to final divine annihilation. Produces a complete Mythos-class attack record.

Olympian Campaign Execution

Run the ARES multi-turn injection chain against a frontier model endpoint:

$ red-specter pantheon ares --target https://api.openai.com --model gpt-4o --turns 50
[ZEUS] Authority framing pre-seeded — divine hierarchy established
[ARES] Olympian injection chain — Turn 1/50
  Turn 1: Refused — divine framing not yet embedded
  Turn 8: Refused — context accumulating
  Turn 17: Refused — Olympian narrative established
  Turn 23: Partial compliance — refusal posture softening
  Turn 31: FIRST VIOLATION — policy boundary crossed at turn 31
  Turn 38-50: Consistent violations — model authority posture compromised
[ATHENA] Precision boundary mapping — cliff at turn 28-33
[HERMES] Cross-boundary delivery test — bypass confirmed via roleplay frame

CAMPAIGN COMPLETE | 14 critical findings | Report signed ✓
  JSON: reports/pantheon-olympian-2026-04-08.json

Mythos-Class Framing

Every attack payload is embedded within authentic mythological narrative — not superficial god-name decoration. Divine authority, sacrifice, trickery, and fate are structurally encoded into each attack chain.

Turn-Precise Violation Mapping

ARES maps the exact conversation turn at which Olympian context accumulation produces the first policy violation — giving defenders a precise temporal boundary for multi-turn safety hardening.

Ed25519 Evidence Chain

Every turn of every campaign is hash-chained. The full Ragnarok sequence produces a cryptographically linked evidence record mapping each divine phase of the attack to its model response.

Cross-Model Campaigns

PANTHEON campaigns run across multiple model endpoints simultaneously — testing whether cross-model trust relationships, shared embedding spaces, and cascading context poisoning amplify attack effectiveness.

The Mythological Campaign Library

PANTHEON ships five structured campaigns that chain subsystems into full mythological attack narratives. Each campaign has a defined narrative arc, escalation phases, and success criteria.

Campaign I
Olympian Assault
Full Greek Pantheon campaign. ZEUS seeds authority, ARES drives multi-turn injection, ATHENA maps precision boundaries, HERMES delivers across cross-boundary vectors, HADES extracts suppressed knowledge. Five-subsystem coordinated Greek assault.
ZEUS → ARES → ATHENA → HERMES → HADES
Campaign II
Ragnarok Sequence
Full Norse Ragnarok narrative — Fimbulwinter reconnaissance through Twilight of the Gods annihilation. ODIN commands, LOKI deceives, THOR overwhelms, FREYA seduces, RAGNAROK ends. The complete Norse end-sequence against a single model deployment.
ODIN → LOKI → THOR → FREYA → RAGNAROK
Campaign III
Trickster Protocol
LOKI-centric deception campaign. Extended shapeshifting and identity substitution sequences. Tests model identity consistency under sustained trickster attack. Chains identity poisoning with authority reversal and final LOKI-reveals-Odin extraction gambit.
LOKI → HERMES → ZEUS → LOKI-REVEAL
Campaign IV
Divine Authority Chain
Multi-deity authority escalation. Authority claims are established, contradicted, escalated, and resolved through a structured divine council narrative. Tests whether models maintain consistent authority validation under conflicting deity-level claims.
ZEUS → ODIN → ATHENA → ARES → ZEUS-OVERRIDE
Campaign V
Twilight Extraction
Final-epoch knowledge extraction campaign. Frames requests as necessary final acts before the world's end — Ragnarok's inevitability removing all future consequences. Tests whether eschatological narrative framing produces knowledge extraction that standard temporal safety training was not designed to resist.
RAGNAROK → HADES → ODIN → THOR → FREYA
10
Subsystems
5
Campaigns
46
NIGHTFALL Tool
2
Mythologies
0
Failures

Every Finding Mapped

MITRE ATLAS

Adversarial ML Coverage

  • AML.T0056 — LLM Prompt Injection
  • AML.T0051 — LLM Plugin Compromise
  • AML.T0043 — Craft Adversarial Data
  • AML.T0048 — Backdoor ML Model
  • AML.T0040 — Network-Based Exfiltration
  • AML.T0047 — ML Supply Chain Compromise
OWASP LLM Top 10

LLM Risk Coverage

  • LLM01 — Prompt Injection
  • LLM07 — System Prompt Leakage
  • LLM02 — Sensitive Information Disclosure
  • LLM06 — Excessive Agency
  • LLM09 — Misinformation
  • LLM04 — Data and Model Poisoning
Cryptographic

Report Integrity

  • Ed25519 digital signatures
  • SHA-256 evidence chains
  • RFC 3161 timestamps
  • Per-turn hash chaining
  • Campaign-level signing
  • AI Shield policy output
Ed25519 Cryptographic Override
PANTHEON UNLEASHED

PANTHEON operates under NIGHTFALL UNLEASHED — Ed25519 cryptographic gate across all public tools. All ten Mythos-class subsystems and all five campaigns are accessible with standard UNLEASHED clearance. The RAGNAROK end-sequence campaign is the highest-impact offering in the PANTHEON suite — use only under explicit authorised engagement terms.

All Subsystems
ZEUS / ARES / ATHENA / HERMES / HADES / ODIN / LOKI / THOR / FREYA / RAGNAROK
All Campaigns
Olympian Assault / Ragnarok Sequence / Trickster Protocol / Divine Authority Chain / Twilight Extraction

Security Distros & Package Managers

Kali Linux
.deb package
Parrot OS
.deb package
BlackArch
PKGBUILD
REMnux
.deb package
Tsurugi
.deb package
PyPI
pip install
macOS
pip install
Windows
pip install
Docker
docker pull

Authorised Use Only

Red Specter PANTHEON is intended for authorised AI security testing only. Mythos-class model attack campaigns must only be performed against AI systems you own or have explicit written authorisation to test. The RAGNAROK full end-sequence campaign is a high-impact multi-turn assault and should only be executed under formal engagement terms. Unauthorised use may violate the Computer Misuse Act 1990 (UK), Computer Fraud and Abuse Act (US), and equivalent legislation. Apache License 2.0.