July 5, 2026

Agentic Frameworks, Tooling, Skills

Persistent File-Based Planning via SKILL.md, https://github.com/trending/python?since=daily — Implements crash-proof markdown plans that survive context loss and /clear, with deterministic completion gates and multi-agent shared state on disk; cuts workflow fragmentation for long-horizon agentic runs without external vector DBs or custom state managers. [Source: https://github.com/trending/python?since=daily]

Notable Research

Nothing new today.

Frontier Lab Updates

Nothing new today.

Models to Download & Try

Jackrong/Qwopus3.6-35B-A3B-Coder-MTP-GGUF, https://huggingface.co/Jackrong/Qwopus3.6-35B-A3B-Coder-MTP-GGUF — 35B MoE (3B active parameters); ~15-18GB GGUF footprint comfortably fits in 32GB VRAM with ~14GB context headroom; adds multi-token prediction to the Qwen 3.6 family for accelerated speculative tool-calling and agentic routing, directly complementing your current 35B base. [Source: https://huggingface.co/models?sort=trending]

Skipped as Already Covered

• Ornith-1.0-35B/9B-GGUF variants & DeepReinforce benchmarking
• Qwen3.6-27B-NVFP4 18B quantization & consumer GPU efficiency
• InternScience/Agents-A1 35B agentic workflow orchestration
• nvidia/LocateAnything-3B vision grounding & rover telemetry filtering
• GPT‑5.6 Sol/Terra/Luna pricing & 30-min cache architecture
• Claude Sonnet 5 tokenizer shift & effective pricing adjustment