The leaked Anthropic reasoning protocol. Running locally. Zero-drift coding with Strict Write Discipline and adaptive Claude Opus 4.7 thinking. Mythos
Outside-in replication of Anthropic's Mythos Preview / Project Glasswing — open-source agentic vulnerability-discovery scaffold on Claude Opus 4.7. Eight-phase sink-guided pipeline, ~$1/run, OSS self-scan and coordinated disclosure.
🚨 The leaked source code of Anthropic's Claude Code. Explore 512k+ lines of TypeScript revealing multi-agent swarms, MCP integrations, Bun runtime, and a React/Ink terminal UI.
Visualization dashboard for parsing Mythos Router MEMORY.md files and SWD protocols on localhost
AI for Anti-Malware Research
OpenMythos Claude Skill: a lightweight add-on that makes Claude act like a senior engineer who deeply understands Kye Gomez’s PyTorch RDT repo. It knows architecture, invariants, patterns, and debugging workflows—helping you reason, extend, and troubleshoot without modifying or redistributing the code.
A finish-first agent loop for Claude Opus 4.7 — plans, executes, verifies, and ships.
AI-Tadpole-OS a local-first development platform for running autonomous AI agent teams on your own hardware. Coordinate parallel workflows, maintain full data privacy, and oversee every decision — no cloud required.
A curated list of autonomous AI agent frameworks, tools, benchmarks, and papers
🔧 Recurrent-Depth Transformer Research Lab — LTI-stable looped inference, switchable MLA/GQA attention, MoE routing & adaptive halting (ACT). Independent research, not affiliated with Anthropic.
Explicit-state recurrent synthesis: deterministic packet compiler for AI agent runs. Takes raw subagent output and produces a schema-validated, hash-provenanced next-pass packet that Prime reads instead of raw subagent prose.
A theoretical reconstruction of the Claude Mythos architecture, built from first principles using the available research literature.
Claude Mythos Preview CLI (Entertainment Purposes)
Resources for the Mythos homebrew script for the game Blood on the Clocktower.
Great Seal of the United States of America - Art
Pattern-based scaffold for systematic AI agent work, with honest capability boundaries. Inspired by observed Claude Mythos Preview behaviors.
Add a description, image, and links to the mythos topic page so that developers can more easily learn about it.
To associate your repository with the mythos topic, visit your repo's landing page and select "manage topics."