蒸餾李宏毅老師的skill，結合Karpathy的LLM，Fable 5和GPT6 加持 以及 本人親自訪談
-
Updated
Aug 11, 2026 - HTML
#
fable5
Here are 53 public repositories matching this topic...
Long Horizon Terminal Benchmark with Dense Reward Grading
minimax claude terminal-bench long-horizon-tasks long-horizon-agents hy3 deepseek-v4-pro fable5 glm-52 grok-45 gpt-56-sol kimi-k3 long-horizon-terminal-bench long-horizon-harness
-
Updated
Aug 27, 2026 - Python
Claude Fable Ultimate
-
Updated
Sep 2, 2026
Make Opus 4.8 (or any Claude model) work like Claude Fable 5 — a Claude Code skill + guard hooks (plan gate, model ceiling, per-task enforcement) for Fable-5-grade discipline: plan, self-verify, and route sub-agents, without Fable 5.
developer-tools opus ai-agents claude llm prompt-engineering anthropic agentic ai-coding agent-orchestration claude-code coding-agent subagents agent-hooks claude-skills claude-fable-5 fable-5 fable5
-
Updated
Jul 15, 2026 - Python
Claude Fable 5.1 prompting guide as Claude Code skills (KO/EN/ZH). Output quality goes up on Opus 5 and Sonnet 5 too.
ai-agents claude prompt-engineering prompting anthropic claude-code claude-code-plugin claude-skills fable-5 fable5
-
Updated
Sep 7, 2026 - Shell
Eco mode for Claude Code. /eco: -31% to -73% output tokens with critical findings intact; /eco-max: up to -75% with lowered effort. Measured hardest on Claude Fable 5 (fable5), deep-studied on Sonnet 5, works on Opus 4.8 too. We publish our negative results. 82 raw benchmark runs.
tokens developer-tools ai-agents claude cost-optimization llm cost-reduction anthropic claude-ai claude-opus claude-sonnet claude-code token-optimization claude-code-plugin claude-skills claude-fable-5 fable-5 fable5
-
Updated
Aug 16, 2026 - JavaScript
Proactive 5h usage guard for Claude Code (macOS). Pause before extra wallet on long batch/subagent runs. Account-level window visibility. /usage-guard skill + daemon. Not affiliated with Anthropic.
macos workflow automation skills fable opus claude anthropic claude-desktop agentic-ai claude-code claude-skills skill-md usage-limits usage-limits-watcher fable-5 fable5
-
Updated
Jul 29, 2026 - Python
AI reasoning skills distilled from 4,665 real Claude Fable 5 chain-of-thought traces. Mathematically tuned. Grade A (100%) emulation accuracy against MiniMax M3.
ai cursor minimax reasoning claude cot llm prompt-engineering chain-of-thought anthropic agent-skills claude-code coding-agent skill-md fable-5 fable5
-
Updated
Jun 16, 2026 - JavaScript
oh-my-fable — Fable 5's way of working a long task (plan first, self-correct, never lose the thread), as a model-agnostic agent harness. The run lives in one serializable RunContext, checkpointed every step, so a crash is a pause. Zero deps, deterministically testable.
agent reflection typescript planner checkpoint zero-dependency resumable reasoning mindset model-agnostic crash-recovery agent-framework self-correcting autonomous-agent ai-agent llm anthropic plan-and-execute fable-5 fable5
-
Updated
Sep 4, 2026 - TypeScript
Reverse-engineer the RL reward an LLM was trained on — from black-box agentic coding behavior alone. Tested on Opus 4.8, Fable 5, GPT-5.5.
benchmark machine-learning reinforcement-learning gpt inverse-reinforcement-learning interpretability ai-agents claude ai-alignment large-language-models llm rlhf mechanistic-interpretability llm-evaluation coding-agents agentic-ai reward-modeling swe-bench reward-hacking fable5
-
Updated
Sep 5, 2026 - Python
Offline, headless pixel-art editor for creating, inspecting, and exporting layered animations through a CLI or MCP. Entirely AI-generated.
-
Updated
Sep 6, 2026 - Rust
A complete breakdown of the Fable 5 constitutional AI jailbreak technique for Claude 4.8
research cybersecurity red-team claude ai-security llm-security constitutional-ai ai-jailbreak fable5
-
Updated
Aug 16, 2026 - HTML
Access Claude Code and Fable 5 through this open source cross-platform desktop client.
claude-alternative claude-code- claude-code-themes claude-desktop-extentions github-claude-code claude-code-tasks everything-claude-code claude-desktop-app claude-code-source-code claude-design-free claude-code-free-helper claude-fable fable5
-
Updated
Sep 8, 2026 - C#
After distilling Fable 5, the output is an indexed CLAUDE.md + on-demand rules + skills, an executable specification designed for weaker model sessions.
-
Updated
Aug 31, 2026 - Python
A curated monorepo of interactive web experiments, 3D flight simulators, physics engines, web OS emulators, and creative tools.
-
Updated
Aug 3, 2026 - JavaScript
Make the model you already use work more like a frontier model with better planning, persistent context, skills, hooks, failure handling, and verification
agent typescript ai bun claude ai-tools llm prompt-engineering anthropic system-prompt gemini-cli agentic-ai claude-code mythos-router fable5 deepseek-harness dsh-plugin cordis-plugin
-
Updated
Sep 9, 2026 - TypeScript
OpenAI Codex plugin for evidence-first code review, audits, fact checks, run contracts, and bounded subagent workflows.
static-analysis audit developer-tools code-review software-engineering software-quality fact-checking codex ecf fact-check openai-codex ai-code-review agent-skills subagents ai-audit codex-plugin fable5
-
Updated
Aug 21, 2026 - JavaScript
Fable 5 성능과 가격에 충격 먹은 사람이 GPT-5.6으로 깎아 만든 Claude,GPT,Gemini 용 md.
-
Updated
Aug 23, 2026
Add this topic to your repo
To associate your repository with the fable5 topic, visit your repo's landing page and select "manage topics."